- Email: [email protected]
Ultrasonographic and endocrine evaluation of ovarian function among Norplant® implants users with regular menses
ELSEVIER
Ultrasonographic and Endocrine Evaluation of Ovarian Function Among Norplant@ Implants Users with Regular Menses Francisco
Alvarez, Vivian Brache, Anibal Faundes,*
With the objective of evaluating the ovulatory function among long-term Norplant implants users with regular menstrual cycles, we undertook this prospective study including 11 Norplant implants usersand 11 control women who were not using hormonal methods of fertility control. Exposed and unexposed women had had at least three regular menstrual cycles preceding enrollment. All women were followed during one menstrual cycle by serial vaginal ultrasound and estradiol (EJ, progesterone (P), LH, and FSH measurements. Three Norplant implants users ovulated, three had luteinization of an unmptured follicle (LUF), three had persistent follicle growth up to a mean of 33 mm without rupture, and two had no follicular development beyond 16 mm. Ten of the controls had normal ovulation and one had LUF. Mean peak LH and FSH among Norplant implants users who ovulated were three- to four-fold lower than among controls. Although usersof Norplant implants with regular cycles frequently have luteal activity, the results of this study suggest that elevation of P during the second half of the cycle doesnot necessarily indicate ovulation has occurred and may frequently be associated with the presenceof Iuteinized unruptured follicle. When ovulation occurs, there are usually abnormal hormone levels (low LHIFSH peak, low progesterone) which may also contribute to the contraceptive effect of Norplant implants. 0 1996 Elsevier Science Inc. All rights reserved. CONTRACEPTION 1996~54:
275-279 Norplant implants, luteinized licle, persistent follicle, anovulation
KEY WORDS:
unruptured fol-
Biomedical Research Department-Profamilia, Santo Domingo, Dominican Republic; and *Department of Obstetrics and Gynecology, Universidade Estadual de Camoinas. Camoinas. Sao Paula. Brazil. Name and address for correspondence: Dr. Francisco Alvarez-Sanchez, Profamilia, P.O. Box 1053. Santo Domingo, Dominican Republic. Tel: (809) 6818253; Fax: (809) 681-8357; E-mail, [email protected] Submitted for publication April 12, 1996 Revised July 5, 1996 Accepted for publication July 9, 1996 aNorplant is the registered trademark of the Population Council, Inc., for levonorgestrel contraceptive subdermal implants.
0 1996 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
Ana Sofia Tejada,
and Francisco
Thevenin
Introduction he Norplant contraceptive system (Leiras Helsinki, Finland] has proven to be a Medical, very effective contraceptive method, although the mechanism for its antifertility effect is not perfectly understood.lf’ Several studies suggested that anovulation is frequent during the initial year of use, coinciding with frequent menstrual disorders, but anovulation ceases to be the main mechanism of contraceptive action when menses become regular in later years of use. Luteal activity has been found in over half of cycles among Norplant implants users with regular menses, suggesting that the main mechanisms of action may be other than anovulation, possibly related to sperm migration blockage by a progestational cervical mucus.3-6 In earlier studies of pituitary-ovarian function during use of Nor-plant implants, we found a consistent inhibition of the normal midcycle gonadotropin surge, with luteinizing hormone (LH) and follicle stimulating hormone (FSH) peak levels consistently below the range observed among controls, accompanied by progesterone (P) serum levels significantly lower than in normal subjects. Based on these findings, we postulated that more subtle disturbances of ovarian function are also an important mechanism of contraceptive action of the implants, and that some of the cycles with luteal activity probably corresponded to luteinized unruptured follicles (LUF) instead of ovulation.7~8 In order to further understand this mechanism, we followed a group of 12 regularly menstruating Norplant implants users and 12 normal controls, for one menstrual cycle, with serial vaginal ultrasonography and blood sampling for estradiol (E,), P, LH and FSH.
T
Materials
and Methods
This study was undertaken at the Department of Biomedical Research of PROFAMILIA, in Santo DomISSN OOlO-7824/96/$15.00 PII SOOIO-7824(96)00179-5
276
Alvarez et al.
Contraception 1996;54:275-279
ingo, Dominican Republic, after approval of the protocol by the Institutional Review Board of our institution, and after obtaining written informed consent from all participants. Twelve volunteers were recruited among the general population of our family planning clinic who had been using Norplant subderma1 contraceptive implants for 2-5 years (n = 3, 3, 4, and 2, respectively). Twelve women, non-users of hormonal contraceptives, were selected as controls. Both groups were required to have at least three regular menstrual cycles, preceding enrollment. One of the Nor-plant implants users did not follow the visit schedule and was excluded due to incomplete information. One of the controls with an abnormally prolonged cycle of 45 days discontinued sampling, and was therefore also excluded from analysis. Vaginal ultrasound and blood sampling were initiated on day 5-7 of the cycle and continued three times per week until a dominant follicle of 2 15 mm was observed. From then on, venipuncture was performed daily, and sonography performed three times per week, until follicle rupture and disappearance was documented, or until the follicle reached a diameter of 825 mm. Therafter, both blood sampling and sonography were performed twice per week until initiation of menses. If no dominant follicle was observed, blood sampling and sonography were performed three times a week for the first two weeks of observation, and then reduced to twice a week until menses. Vaginal ultrasounds were performed with a realtime scanner Ultramark 4 (Advanced Technology Laboratories, Inc., Bothell, WA) with a 5.0 MHz vaginal transducer. Estradiol, P, LH, and FSH were measured in the blood samples. Estradiol and P were analyzed using a (RIA) (DPC, Los solid-phase i2? radioimmunoassay Angeles, CA). Interassay coefficients of variation were 6.52 and 6.06%, respectively. LH and FSH was determined by a liquid phase double antibody “‘1 RIA provided by the World Health Organization Matched Reagent Programme (Geneva, Switzerland), standardized against IRP 68/40 and with a coefficient of variation of 13.7% and 16.4%, respectively. Quality control was ensured both by the External Quality Table
Assessment Scheme for Reproductive Hormones of the World Health Organization and by internal quality control with commercial sera. In each assay, all samples from both one Norplant implants user and one control were run together to reduce the effect of interassay variability. Statistical differences between groups and subgroups were evaluated by comparison of the means using Student’s t-test. The 95% confidence intervals for proportions were calculated.
Results The sonographic follow-up of follicular development allowed us to distinguish four patterns among Norplant implants users: ovulation, luteinized unruptured follicle, persistent follicle, and no growth of a follicle > 16 mm (Table 1). Nine of the eleven users of the Norplant implants developed follicles larger than 16 mm diameter. Three of these 11 ovulated (27%), as documented by follicle rupture and disappearance, after reaching a mean diameter of 18.0, 25.6, and 26 mm, followed by a rise in P serum levels between 6.4 and 10.8 ng/mL (Tables 1 and 2). Three subjects had continuous growth of a follicle (27%), which reached a mean diameter between 28 and 58 mm, but never showed signs of rupture. Instead, these follicles developed internal echoes, characteristic of a LUF (Figure 1). These three subjects had a lower peak progesterone level, ranging between 2.6 and 5.2 ng/mL (Table 2). The mean highest progesterone level for the subgroup with LUF, 4.0 ng/mL, was significantly lower than the highest level in the subgroup who had ovulated, 9.0 ng/mL (p ~0.05); however, both of these subgroups who used Norplant implants had significantly lower mean peak progesterone levels than the ovulatory controls, 15.1 ng/mL (p
1. Ultrasonographic observation of follicular development among Norplant implants userswith regular mensesand
controls Norplant Implants Classification Ovulation Luteinized unruptured follicle Persistent follicle No follicle above 16 mm
Controls
N
%
95% CI
N
%
95% CI
3/11 3/11 3/11 2/11
27 27 27 18
6-61 6-61 6-61 2-52
10/l 1 l/11 o/11 o/11
91 9 0 0
59-100 o-41
Contraception 1996;54:275-279
Table
Ovarian
Function
in Norplant@
Implants
277
Users
2. Ultrasonographic and hormonal results among Norplant implants users with regular mensesand controls Norplant Implants Users Controls Ovulatory
N Maximum follicular diameter (mm)
FSH peak (IU/L)
LUF
Persistent Follicle
10
1
3
3
3
29.6
SD X
10.0
23.2 4.5 15.7 5.5 2.3 0.4 291.6 73.8 111.9 27.5 9.0 2.3
47.3 16.8 14.5 9.2 2.3 1.4 333.9 26.9 349.7 59.6 4.0 1.3
32.7 2.0 12.0 7.6 2.0 1.9 237.0 113.5 592.1 144.2 0.6 0.1
X
SD Midcycle E, (pg/mL)
X
Highest E, in second half of cycle (pg/mL)
SD x SD
Highest P in secondhalf of cycle (ng/mL)
Ovulatory
20.3 2.8 57.2 23.3
fzD
LH peak (W/L)
LUF
X
SD
22.1 2.0
4.3
271.5 79.1 183.5
268.8 249.5
53.5
15.1 2.7
8.8
showed limited follicular development to a diameter E, levels not greater than 16 mm, with fluctuating below 100 pg/mL and P levels continuously below 1 ng/mL throughout the cycle (Table 1). Ten of the eleven controls (91%) had a normal ovulatory endocrine profile and follicle rupture was observed within two days after reaching a diameter between 15.3 and 24 mm, followed by a rise in progesterone ranging between 12.5 and 19.9 ng/mL in the mid-luteal phase (Tables 1 and 2). There was one case of LUF, which had an um-uptured follicle that developed up to 29.6 mm, a blunted LH peak of 22.1 IU/L, and highest P level of 8.8 ng/mL. In none of the Norplant implants users was a FSH peak observed, whereas the mean FSH peak for the control group was 10.0 IU/L (Table 2). The highest LH level in all subgroups of Norplant implants users was also significantly lower than among controls, but not
different within each other. The highest LH level found among Norplant implants users was always lower than the lowest value observed among normal controls. Estradiol levels at midcycle were similar for all subgroups of controls and Norplant implants users, with the exception of the two subjects who had limited follicular development, whose E, levels never exceeded 100 pg/mL. The mean peak E, level on days -8 to -5 preceding menses among Norplant implants users that had evidence of LUF (349.7 pg/mL) was almost twice that of controls (183.5 pg/mL) (p
Figure 1. Sonographic image of a luteinized unruptured follicle (mean diameter 51.0 mm) on day 23 of the cycle in a Norplant implants user.
Figure 2. Sonographic image of a persistent anechogenic follicle (mean diameter 31.6 mm) on day 25 of the cycle in a Norplant implants user.
278
Contraception 1996;54:275-279
Alvarez et al.
Discussion The data presented here confirm our hypothesis that a substantial proportion of cycles with luteal activity among Norplant implants users (three of six cycles; 95% confidence interval of 14-86%, in this series) have a luteinized unruptured follicle. It also confirms the previously described absent or very low gonadotropin surge, decreased progesterone secretion in ovulatory cycles, and the frequent presence of persistent anovulatory follicles during Nor-plant implants use. These results are similar to those reported by Olsson et a1.9, Shoupe et al.” and Shaaban et al.” However, Shaaban et al. described the presence of persistent follicles, which they named the “carryover” phenomenon, in which menses was sometimes initiated during the presence of high E, levels. In our series, menstruation always was initiated with declining and low estradiol levels, in spite of the persistence of a large follicle. In women not receiving continuous progestin treatment, LUF cycles are characterized by reduced LH peak and low luteal progesterone levels.‘2,‘3 The Norplant implants users with LUF had luteal phase progesterone levels even lower than those observed among Nor-plant implants users who ovulated. These women, in turn, had diminished luteal activity in comparison with non-treated controls. The mean LH peak level was not different among Nor-plant implants users with LUF than among those who ovulated. However, we have previously shown and confirmed in this study, that LH peak during use of Norplant implants is always lower than that observed in non-treated women.7,8 Another characteristic of LUF cycles among Norplant implants users was a significantly higher estradiol level during the luteal phase in comparison with those found during ovulatory cycles in women using Norplant implants. This finding has not yet been described as one of the endocrine characteristics of cycles with LUF in non-pharmacological conditions. It is not clear, however, if the elevated E, level is something that has not been reported or has not been observed. For example, the estradiol data presented by Hamilton et al. l2 indicated that there were higher levels of estradiol in the early luteal phase among the LUF subjects than in controls; however, this fact was not commented upon by the authors. In the present study, the one control subject who had a LUF had all the characteristics of this syndrome and also a higher luteal phase serum estradiol level than found in ovulatory cycles. Other authors have indicated that follicular rupture is important for corpus luteum progesterone secretion.14 The lack of follicular rupture may also stimu-
late greater E, production, mimicking at a lower level what is observed in cycles with persistent follicles. Although we cannot generalize from very few cases, it will be interesting if other investigators would investigate this aspect of the LUF syndrome. In a previous publication of endocrine profiles among Nor-plant implants users, we had described a subgroup of women, which we classified as those with “minimal luteal activity,” who had lower progesterone and higher estradiol mid-luteal levels than normal. We postulated that these cases had LUF. The similar data presented here, in confirmed LUF cycles, would seem to reinforce this hypothesis. Thus, observation of the endocrine profile may allow us to presume the morphologic evolution of the follicle unit in a given cycle, at least when continuous administration of low doses of progestogens are used. A large percentage of Nor-plant implants users have menstrual alterations during the first year of use, when anovulation is most common, due to the higher initial release rate of levonorgestrel from the capsules. During longer-term use, a more regular menstrual pattern predominates. We and other authors have shown that when this occurs, follicle development and ovarian function may still not be normal. Ovulation, defined by ultrasound documentation of follicle rupture, occurs in only about 2535% of the cycles and, even in these instances, endocrine disturbances are present. Moreover, the apparently high frequency of LUF in this, as well as in other studies, suggests that the incidence of real ovulation is much lower than the 40-80% per cycle reported in the literature, based on progesterone elevation during the second half of the cycle among long-term Norplant implants users.3-5
References 1. Sivin I. International
experience with Norplant implants and Nolplant implants-2 contraceptives. Stud Fam Plann 1988;19:81. 2. Grubb GS, Moore D, Anderson NG. Pre-introductory clinical trials of Norplant implants: A comparison of seventeen countries experience. Contraception 1995; 52:287-96. 3. Croxatto HB, Diaz S, Pavez M, Miranda P, Brandeis A.
Plasma progesterone levels during long-term treatment with levonorgestrel Silastic implants. Acta Endocrinol (Copenh) 1992jlOlZ307. 4. Brache V, FarIndes A, Johansson E, Alvarez F. Anovulation, inadequate luteal phase and poor sperm penetration in cervical mucus during prolonged use of Norplant implants. Contraception 1985j31:261. 5. Brache V, Alvarez F, Faundes A, Tejada AS, Cochdn L. Ovarian endocrine function through five years continuous treatment with Norplant subdermal implants. Contraception 1990;41:160-77.
Contraception 1996;54:275-279
6. Croxatto HB, Diaz S, Salvatierra AM, Morales P, EbenspergerC, Brandeis A. Treatment with Norplant subdermal implants inhibits sperm penetration through cervical mucus in vitro. Contraception 1987;36:193. 7. Alvarez F, Brache V, Tejada AS, Fatindes A. Abnormal endocrine profile among women with confirmed or presumed ovulation during long-term Norplant implants use. Contraception 1986;33:11l-9. 8. Faiindes A, Brache V, Tejada AS, Cochon L, Alvarez F. Ovulatory dysfunction during continuous administration of low-dose levonorgestrel by subdermal implants. Fertil Steril 1991;56:27-31. 9. Olsson SE, Bakos 0, Lindgren PG, Odlind V, Wide L. Ovarian function during use of subdermal implants releasing low dosesof levonorgestrel. Brit J Fam Plann 1990;16:88-93. 10. Shoupe D, Horenstein J, Mishell Jr DR, Lacarra M, Medearis A. Characteristics of ovarian follicular devel-
Ovarian
Function
in Norplant@
Implants
Users
277
opment in Norplant implants users. Fertil Steril 1991; 55:766-70. 11. Shaaban MM, Segal S, Salem HT, Ghaneimah SA, Khalifa E-AM, Ahmed A-G. Sonographic assessmentof ovarian and endometrial changesduring long-term Norplant implants use and their correlation with hormonal levels. Fertil Steril 1993;59:998-1002. 12. Hamilton CJCM, Wetzels LCG, Evers JLH, Hoogland HJ, Muijtjens A, de Haan J. Follicle growth curves and hormonal patterns in patients with the luteinized unruptured follicle syndrome. Fertil Steril 1985;43:541-8. 13. Hamilton CJCM, Evers JLH, de Haan J. Ovulatory disturbances in patients with luteal insufficiency. Clin Endocrinol (Oxford) 1987;26:129-36. 14. Hamilton MPR, Fleming R, Coutts JRT, Macnaughton MC, Whitfield CR. Luteal cysts and unexplained infertility: Biochemical and ultrasonic evaluation. Fertil Steril 1990;54:32-7.
Ultrasonographic and Endocrine Evaluation of Ovarian Function Among Norplant@ Implants Users with Regular Menses Francisco
Alvarez, Vivian Brache, Anibal Faundes,*
With the objective of evaluating the ovulatory function among long-term Norplant implants users with regular menstrual cycles, we undertook this prospective study including 11 Norplant implants usersand 11 control women who were not using hormonal methods of fertility control. Exposed and unexposed women had had at least three regular menstrual cycles preceding enrollment. All women were followed during one menstrual cycle by serial vaginal ultrasound and estradiol (EJ, progesterone (P), LH, and FSH measurements. Three Norplant implants users ovulated, three had luteinization of an unmptured follicle (LUF), three had persistent follicle growth up to a mean of 33 mm without rupture, and two had no follicular development beyond 16 mm. Ten of the controls had normal ovulation and one had LUF. Mean peak LH and FSH among Norplant implants users who ovulated were three- to four-fold lower than among controls. Although usersof Norplant implants with regular cycles frequently have luteal activity, the results of this study suggest that elevation of P during the second half of the cycle doesnot necessarily indicate ovulation has occurred and may frequently be associated with the presenceof Iuteinized unruptured follicle. When ovulation occurs, there are usually abnormal hormone levels (low LHIFSH peak, low progesterone) which may also contribute to the contraceptive effect of Norplant implants. 0 1996 Elsevier Science Inc. All rights reserved. CONTRACEPTION 1996~54:
275-279 Norplant implants, luteinized licle, persistent follicle, anovulation
KEY WORDS:
unruptured fol-
Biomedical Research Department-Profamilia, Santo Domingo, Dominican Republic; and *Department of Obstetrics and Gynecology, Universidade Estadual de Camoinas. Camoinas. Sao Paula. Brazil. Name and address for correspondence: Dr. Francisco Alvarez-Sanchez, Profamilia, P.O. Box 1053. Santo Domingo, Dominican Republic. Tel: (809) 6818253; Fax: (809) 681-8357; E-mail, [email protected] Submitted for publication April 12, 1996 Revised July 5, 1996 Accepted for publication July 9, 1996 aNorplant is the registered trademark of the Population Council, Inc., for levonorgestrel contraceptive subdermal implants.
0 1996 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
Ana Sofia Tejada,
and Francisco
Thevenin
Introduction he Norplant contraceptive system (Leiras Helsinki, Finland] has proven to be a Medical, very effective contraceptive method, although the mechanism for its antifertility effect is not perfectly understood.lf’ Several studies suggested that anovulation is frequent during the initial year of use, coinciding with frequent menstrual disorders, but anovulation ceases to be the main mechanism of contraceptive action when menses become regular in later years of use. Luteal activity has been found in over half of cycles among Norplant implants users with regular menses, suggesting that the main mechanisms of action may be other than anovulation, possibly related to sperm migration blockage by a progestational cervical mucus.3-6 In earlier studies of pituitary-ovarian function during use of Nor-plant implants, we found a consistent inhibition of the normal midcycle gonadotropin surge, with luteinizing hormone (LH) and follicle stimulating hormone (FSH) peak levels consistently below the range observed among controls, accompanied by progesterone (P) serum levels significantly lower than in normal subjects. Based on these findings, we postulated that more subtle disturbances of ovarian function are also an important mechanism of contraceptive action of the implants, and that some of the cycles with luteal activity probably corresponded to luteinized unruptured follicles (LUF) instead of ovulation.7~8 In order to further understand this mechanism, we followed a group of 12 regularly menstruating Norplant implants users and 12 normal controls, for one menstrual cycle, with serial vaginal ultrasonography and blood sampling for estradiol (E,), P, LH and FSH.
T
Materials
and Methods
This study was undertaken at the Department of Biomedical Research of PROFAMILIA, in Santo DomISSN OOlO-7824/96/$15.00 PII SOOIO-7824(96)00179-5
276
Alvarez et al.
Contraception 1996;54:275-279
ingo, Dominican Republic, after approval of the protocol by the Institutional Review Board of our institution, and after obtaining written informed consent from all participants. Twelve volunteers were recruited among the general population of our family planning clinic who had been using Norplant subderma1 contraceptive implants for 2-5 years (n = 3, 3, 4, and 2, respectively). Twelve women, non-users of hormonal contraceptives, were selected as controls. Both groups were required to have at least three regular menstrual cycles, preceding enrollment. One of the Nor-plant implants users did not follow the visit schedule and was excluded due to incomplete information. One of the controls with an abnormally prolonged cycle of 45 days discontinued sampling, and was therefore also excluded from analysis. Vaginal ultrasound and blood sampling were initiated on day 5-7 of the cycle and continued three times per week until a dominant follicle of 2 15 mm was observed. From then on, venipuncture was performed daily, and sonography performed three times per week, until follicle rupture and disappearance was documented, or until the follicle reached a diameter of 825 mm. Therafter, both blood sampling and sonography were performed twice per week until initiation of menses. If no dominant follicle was observed, blood sampling and sonography were performed three times a week for the first two weeks of observation, and then reduced to twice a week until menses. Vaginal ultrasounds were performed with a realtime scanner Ultramark 4 (Advanced Technology Laboratories, Inc., Bothell, WA) with a 5.0 MHz vaginal transducer. Estradiol, P, LH, and FSH were measured in the blood samples. Estradiol and P were analyzed using a (RIA) (DPC, Los solid-phase i2? radioimmunoassay Angeles, CA). Interassay coefficients of variation were 6.52 and 6.06%, respectively. LH and FSH was determined by a liquid phase double antibody “‘1 RIA provided by the World Health Organization Matched Reagent Programme (Geneva, Switzerland), standardized against IRP 68/40 and with a coefficient of variation of 13.7% and 16.4%, respectively. Quality control was ensured both by the External Quality Table
Assessment Scheme for Reproductive Hormones of the World Health Organization and by internal quality control with commercial sera. In each assay, all samples from both one Norplant implants user and one control were run together to reduce the effect of interassay variability. Statistical differences between groups and subgroups were evaluated by comparison of the means using Student’s t-test. The 95% confidence intervals for proportions were calculated.
Results The sonographic follow-up of follicular development allowed us to distinguish four patterns among Norplant implants users: ovulation, luteinized unruptured follicle, persistent follicle, and no growth of a follicle > 16 mm (Table 1). Nine of the eleven users of the Norplant implants developed follicles larger than 16 mm diameter. Three of these 11 ovulated (27%), as documented by follicle rupture and disappearance, after reaching a mean diameter of 18.0, 25.6, and 26 mm, followed by a rise in P serum levels between 6.4 and 10.8 ng/mL (Tables 1 and 2). Three subjects had continuous growth of a follicle (27%), which reached a mean diameter between 28 and 58 mm, but never showed signs of rupture. Instead, these follicles developed internal echoes, characteristic of a LUF (Figure 1). These three subjects had a lower peak progesterone level, ranging between 2.6 and 5.2 ng/mL (Table 2). The mean highest progesterone level for the subgroup with LUF, 4.0 ng/mL, was significantly lower than the highest level in the subgroup who had ovulated, 9.0 ng/mL (p ~0.05); however, both of these subgroups who used Norplant implants had significantly lower mean peak progesterone levels than the ovulatory controls, 15.1 ng/mL (p
1. Ultrasonographic observation of follicular development among Norplant implants userswith regular mensesand
controls Norplant Implants Classification Ovulation Luteinized unruptured follicle Persistent follicle No follicle above 16 mm
Controls
N
%
95% CI
N
%
95% CI
3/11 3/11 3/11 2/11
27 27 27 18
6-61 6-61 6-61 2-52
10/l 1 l/11 o/11 o/11
91 9 0 0
59-100 o-41
Contraception 1996;54:275-279
Table
Ovarian
Function
in Norplant@
Implants
277
Users
2. Ultrasonographic and hormonal results among Norplant implants users with regular mensesand controls Norplant Implants Users Controls Ovulatory
N Maximum follicular diameter (mm)
FSH peak (IU/L)
LUF
Persistent Follicle
10
1
3
3
3
29.6
SD X
10.0
23.2 4.5 15.7 5.5 2.3 0.4 291.6 73.8 111.9 27.5 9.0 2.3
47.3 16.8 14.5 9.2 2.3 1.4 333.9 26.9 349.7 59.6 4.0 1.3
32.7 2.0 12.0 7.6 2.0 1.9 237.0 113.5 592.1 144.2 0.6 0.1
X
SD Midcycle E, (pg/mL)
X
Highest E, in second half of cycle (pg/mL)
SD x SD
Highest P in secondhalf of cycle (ng/mL)
Ovulatory
20.3 2.8 57.2 23.3
fzD
LH peak (W/L)
LUF
X
SD
22.1 2.0
4.3
271.5 79.1 183.5
268.8 249.5
53.5
15.1 2.7
8.8
showed limited follicular development to a diameter E, levels not greater than 16 mm, with fluctuating below 100 pg/mL and P levels continuously below 1 ng/mL throughout the cycle (Table 1). Ten of the eleven controls (91%) had a normal ovulatory endocrine profile and follicle rupture was observed within two days after reaching a diameter between 15.3 and 24 mm, followed by a rise in progesterone ranging between 12.5 and 19.9 ng/mL in the mid-luteal phase (Tables 1 and 2). There was one case of LUF, which had an um-uptured follicle that developed up to 29.6 mm, a blunted LH peak of 22.1 IU/L, and highest P level of 8.8 ng/mL. In none of the Norplant implants users was a FSH peak observed, whereas the mean FSH peak for the control group was 10.0 IU/L (Table 2). The highest LH level in all subgroups of Norplant implants users was also significantly lower than among controls, but not
different within each other. The highest LH level found among Norplant implants users was always lower than the lowest value observed among normal controls. Estradiol levels at midcycle were similar for all subgroups of controls and Norplant implants users, with the exception of the two subjects who had limited follicular development, whose E, levels never exceeded 100 pg/mL. The mean peak E, level on days -8 to -5 preceding menses among Norplant implants users that had evidence of LUF (349.7 pg/mL) was almost twice that of controls (183.5 pg/mL) (p
Figure 1. Sonographic image of a luteinized unruptured follicle (mean diameter 51.0 mm) on day 23 of the cycle in a Norplant implants user.
Figure 2. Sonographic image of a persistent anechogenic follicle (mean diameter 31.6 mm) on day 25 of the cycle in a Norplant implants user.
278
Contraception 1996;54:275-279
Alvarez et al.
Discussion The data presented here confirm our hypothesis that a substantial proportion of cycles with luteal activity among Norplant implants users (three of six cycles; 95% confidence interval of 14-86%, in this series) have a luteinized unruptured follicle. It also confirms the previously described absent or very low gonadotropin surge, decreased progesterone secretion in ovulatory cycles, and the frequent presence of persistent anovulatory follicles during Nor-plant implants use. These results are similar to those reported by Olsson et a1.9, Shoupe et al.” and Shaaban et al.” However, Shaaban et al. described the presence of persistent follicles, which they named the “carryover” phenomenon, in which menses was sometimes initiated during the presence of high E, levels. In our series, menstruation always was initiated with declining and low estradiol levels, in spite of the persistence of a large follicle. In women not receiving continuous progestin treatment, LUF cycles are characterized by reduced LH peak and low luteal progesterone levels.‘2,‘3 The Norplant implants users with LUF had luteal phase progesterone levels even lower than those observed among Nor-plant implants users who ovulated. These women, in turn, had diminished luteal activity in comparison with non-treated controls. The mean LH peak level was not different among Nor-plant implants users with LUF than among those who ovulated. However, we have previously shown and confirmed in this study, that LH peak during use of Norplant implants is always lower than that observed in non-treated women.7,8 Another characteristic of LUF cycles among Norplant implants users was a significantly higher estradiol level during the luteal phase in comparison with those found during ovulatory cycles in women using Norplant implants. This finding has not yet been described as one of the endocrine characteristics of cycles with LUF in non-pharmacological conditions. It is not clear, however, if the elevated E, level is something that has not been reported or has not been observed. For example, the estradiol data presented by Hamilton et al. l2 indicated that there were higher levels of estradiol in the early luteal phase among the LUF subjects than in controls; however, this fact was not commented upon by the authors. In the present study, the one control subject who had a LUF had all the characteristics of this syndrome and also a higher luteal phase serum estradiol level than found in ovulatory cycles. Other authors have indicated that follicular rupture is important for corpus luteum progesterone secretion.14 The lack of follicular rupture may also stimu-
late greater E, production, mimicking at a lower level what is observed in cycles with persistent follicles. Although we cannot generalize from very few cases, it will be interesting if other investigators would investigate this aspect of the LUF syndrome. In a previous publication of endocrine profiles among Nor-plant implants users, we had described a subgroup of women, which we classified as those with “minimal luteal activity,” who had lower progesterone and higher estradiol mid-luteal levels than normal. We postulated that these cases had LUF. The similar data presented here, in confirmed LUF cycles, would seem to reinforce this hypothesis. Thus, observation of the endocrine profile may allow us to presume the morphologic evolution of the follicle unit in a given cycle, at least when continuous administration of low doses of progestogens are used. A large percentage of Nor-plant implants users have menstrual alterations during the first year of use, when anovulation is most common, due to the higher initial release rate of levonorgestrel from the capsules. During longer-term use, a more regular menstrual pattern predominates. We and other authors have shown that when this occurs, follicle development and ovarian function may still not be normal. Ovulation, defined by ultrasound documentation of follicle rupture, occurs in only about 2535% of the cycles and, even in these instances, endocrine disturbances are present. Moreover, the apparently high frequency of LUF in this, as well as in other studies, suggests that the incidence of real ovulation is much lower than the 40-80% per cycle reported in the literature, based on progesterone elevation during the second half of the cycle among long-term Norplant implants users.3-5
References 1. Sivin I. International
experience with Norplant implants and Nolplant implants-2 contraceptives. Stud Fam Plann 1988;19:81. 2. Grubb GS, Moore D, Anderson NG. Pre-introductory clinical trials of Norplant implants: A comparison of seventeen countries experience. Contraception 1995; 52:287-96. 3. Croxatto HB, Diaz S, Pavez M, Miranda P, Brandeis A.
Plasma progesterone levels during long-term treatment with levonorgestrel Silastic implants. Acta Endocrinol (Copenh) 1992jlOlZ307. 4. Brache V, FarIndes A, Johansson E, Alvarez F. Anovulation, inadequate luteal phase and poor sperm penetration in cervical mucus during prolonged use of Norplant implants. Contraception 1985j31:261. 5. Brache V, Alvarez F, Faundes A, Tejada AS, Cochdn L. Ovarian endocrine function through five years continuous treatment with Norplant subdermal implants. Contraception 1990;41:160-77.
Contraception 1996;54:275-279
6. Croxatto HB, Diaz S, Salvatierra AM, Morales P, EbenspergerC, Brandeis A. Treatment with Norplant subdermal implants inhibits sperm penetration through cervical mucus in vitro. Contraception 1987;36:193. 7. Alvarez F, Brache V, Tejada AS, Fatindes A. Abnormal endocrine profile among women with confirmed or presumed ovulation during long-term Norplant implants use. Contraception 1986;33:11l-9. 8. Faiindes A, Brache V, Tejada AS, Cochon L, Alvarez F. Ovulatory dysfunction during continuous administration of low-dose levonorgestrel by subdermal implants. Fertil Steril 1991;56:27-31. 9. Olsson SE, Bakos 0, Lindgren PG, Odlind V, Wide L. Ovarian function during use of subdermal implants releasing low dosesof levonorgestrel. Brit J Fam Plann 1990;16:88-93. 10. Shoupe D, Horenstein J, Mishell Jr DR, Lacarra M, Medearis A. Characteristics of ovarian follicular devel-
Ovarian
Function
in Norplant@
Implants
Users
277
opment in Norplant implants users. Fertil Steril 1991; 55:766-70. 11. Shaaban MM, Segal S, Salem HT, Ghaneimah SA, Khalifa E-AM, Ahmed A-G. Sonographic assessmentof ovarian and endometrial changesduring long-term Norplant implants use and their correlation with hormonal levels. Fertil Steril 1993;59:998-1002. 12. Hamilton CJCM, Wetzels LCG, Evers JLH, Hoogland HJ, Muijtjens A, de Haan J. Follicle growth curves and hormonal patterns in patients with the luteinized unruptured follicle syndrome. Fertil Steril 1985;43:541-8. 13. Hamilton CJCM, Evers JLH, de Haan J. Ovulatory disturbances in patients with luteal insufficiency. Clin Endocrinol (Oxford) 1987;26:129-36. 14. Hamilton MPR, Fleming R, Coutts JRT, Macnaughton MC, Whitfield CR. Luteal cysts and unexplained infertility: Biochemical and ultrasonic evaluation. Fertil Steril 1990;54:32-7.