- Email: [email protected]
Serum levonorgestrel levels and endometrial thickness during extended use of the levonorgestrel-releasing intrauterine system
Contraception 80 (2009) 84 – 89
Original research article
Serum levonorgestrel levels and endometrial thickness during extended use of the levonorgestrel-releasing intrauterine system Maria M. Hidalgo, Creusa Hidalgo-Regina, M. Valeria Bahamondes, Ilza Monteiro, Carlos A. Petta, Luis Bahamondes⁎ Human Reproduction Unit, Department of Obstetrics and Gynecology, School of Medicine, Universidade Estadual de Campinas (UNICAMP), 13084-971 Campinas, Brazil Received 18 June 2008; revised 8 January 2009; accepted 8 January 2009
Abstract Background: The levonorgestrel-releasing intrauterine system (LNG-IUS) is a contraceptive method approved for five years of use. However, there is some evidence that its life span may be longer. The aim of the study was to evaluate serum levonorgestrel (LNG) and estradiol (E2) levels and endometrial thickness every 6 months from 7 to 8 1/2 years after insertion. Study Design: At the end of the approved 5-year life span, no replacement devices were available; therefore, 86 women were allowed to retain the same device for a further 2 years. At the 7-year follow-up visit, the women who consented were again allowed to retain the same device for a further 18 months and were followed-up at six-monthly intervals. At each visit, vaginal sonography was performed to measure endometrial thickness, and a blood sample was taken to measure LNG and E2. Results: Eighty-four months after insertion, 67 women aged 34.3±0.8 years (mean±SEM) (range, 25–49 years) returned for follow-up. Mean±SEM LNG levels decreased from an initial 253±27 pg/mL (range, 86–760) during the first 2 months following insertion to 137±12 (range, 23–393) at 84 months and 119±9 pg/mL (range, 110–129) at 102 months of use (±SEM). At 84 months of use, mean±SEM endometrial thickness was 2.8±0.1 mm, increasing to 3.8±0.5 mm at 102 months of use. The incidence of amenorrhea decreased from 41.8% at 84 months to 31.5% at 102 months of use. No correlation was found between LNG levels and bleeding patterns; however, a weak correlation was found between high body mass index (kg/m2), high weight, and low serum LNG levels. E2 levels were similar to those of the follicular phase of the menstrual cycle of regularly menstruating women. Conclusions: During extended use of the LNG-IUS, serum LNG levels were nearly half those found in the first 2 months of use (Wilcoxon signed rank test); serum E2 levels were normal. Despite the very thin endometrium, menstrual bleeding was reinstated in many cases. At the end of its 5-year life span, there is a window for changing the LNG-IUS, and physicians and users should not be concerned about delaying replacement of the device for a short time beyond the approved life span; however, maintaining the same device long after its approved life span cannot be recommended. © 2009 Elsevier Inc. All rights reserved. Keywords: Levonorgestrel-releasing intrauterine system; Mirena®; Levonorgestrel; Contraception
1. Introduction Long-term contraceptive methods are desired by many women because they provide contraception without the requirement of having to remember to take a daily pill or a monthly or three-monthly injection. Long-term methods include subdermal implants, intrauterine devices and the levonorgestrel-releasing intrauterine system (LNG-IUS). ⁎ Corresponding author. Tel.: +55 19 3289 2856; fax: +55 19 3289 2440. E-mail address: [email protected] (L. Bahamondes). 0010-7824/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2009.01.004
The LNG-IUS is marketed and approved in 123 countries for use as a contraceptive method. One of the effects of the use of the LNG-IUS is amenorrhea caused by the antiproliferative effect of levonorgestrel (LNG) on the endometrium [1–3]. As a consequence of this effect, the device has been approved and is currently in use in many countries as a treatment for idiopathic menorrhagia [4,5] and as endometrial protection during estrogen therapy [6,7]. In addition, this device has been used off-label as a treatment for endometriosis [8,9] and for endometrial hyperplasia [10,11].
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89
The LNG-IUS is approved for 5 years of use during which time the device provides high contraceptive efficacy with few adverse effects [1,12]. Some limited data exist on the device's extended use beyond 5 years and up to 7 years [2,13,14]. The objectives of this study were to evaluate the serum levels of LNG and estradiol (E2) and the endometrial thickness of women who used the LNG-IUS beyond its approved life span of 5 years.
2. Materials and methods The study was conducted at the Human Reproduction Unit, Department of Obstetrics and Gynecology, School of Medical Sciences, Universidade Estadual de Campinas, Campinas, Brazil. The ethical committee of the university approved the study, and all participants signed an informed consent form prior to admission. The study population consisted of a subsample from a larger group of women participating in an acceptability study of the LNG-IUS (Mirena®, Bayer Schering Pharma Oy, Turku, Finland) in Brazil who had their devices inserted between April and September 1998 [3,15]. At the end of the approved 5-year life span of the device, no new devices were available to replace the ones in use; therefore, any woman who wished to do so were allowed to retain the same device for a further 2 years. At the 7-year follow-up visit, the women in the extended use protocol were once again allowed to retain the same device for a further 18 months and were invited to return for follow-up visits at six-monthly intervals. At each visit, the threads of the LNG-IUS were visualized during a gynecological examination. Next, ultrasonography was performed using a 6.0-MHz vaginal probe (Justavision 400, Toshiba, Toshigi-Ken, Japan) to measure double-layer endometrial thickness and to be sure that the LNG-IUS was located in the uterine fundus. A blood sample was then collected, and the serum was separated and stored at −20°C until assayed. Weight for each woman was evaluated at each follow-up visit and body mass index (BMI, kg/m2) was calculated. Also, we reviewed and retrieved the
85
data from the menstrual calendar kept by each woman during the 90 days preceding the interview. According to the World Health Organization [16,17], bleeding was defined as “any bloody vaginal discharge that requires the use of protection such as pads and tampons” and spotting as “any bloody vaginal discharge that is not intense enough to require sanitary protection”. Amenorrhea was defined as no bleeding during the preceding 90 days before the interview and oligomenorrhea as bleeding episodes with more than 35 days interval. Serum samples from women who had insertion of an LNG-IUS in 1998 had been collected at the first scheduled follow-up visit after insertion, and aliquots were kept in the freezer. In this clinic, the first follow-up visit was generally scheduled 30–50 days after insertion. Blood was collected 27–50 days following insertion. Follow-up visits were carried out at six-monthly intervals from 84 months until 102 months after insertion. 2.1. LNG and E2 assay Serum LNG levels were measured using a validated method of high-performance liquid chromatography coupled with tandem mass spectrometry. All procedures were carried out in compliance with the guidelines approved by the Brazilian regulatory authorities (ANVISA) for Good Laboratory Practices. The LNG standards used in the assay were obtained from USP (Rockville, MD, USA). Biological specificity of the method was checked by processing independent serum samples and blank samples obtained from women not using any kind of hormones. Briefly, the bioanalytical assay for the quantification of LNG was developed using an online solid phase extraction method (Symbiosis, Spark Holland, Emmen, The Netherlands). The mass spectrometer (Sciex/Applied Biosystems, model API5000, Toronto, Canada) was equipped with a photo-ionization source (APPI) running in positive ion mode setup in Multiple Reaction Monitoring for the transition m/z 313.3 245.1 for LNG. Toluene was used as solvent at the flow rate of 0.15 mL/min. The lowest limit of quantification of the method was 20 pg/mL, and it was linear over the range
Table 1 Serum LNG levels (pg/mL) and endometrial thickness (mm) at different times after insertion of the LNG-IUS Time after insertion
Days 27–50 Months 84 90 96 102
Age at blood collection
LNG (pg/mL)
Mean±SEM
na
Mean±SEM
33.4±0.71
86
253±27
86–760
34.3±0.75 35.2±1.00 34.4±1.20 41.5±3.50
67 51 25 2b
137±12 130±11 132±18 119±9
23–393 18–470 51–505 110–129
Range
Endometrial thickness (mm)
E2 (pg/mL)
n
n
Mean±SEM
86
99.2±11.4
67 58 34 8b
86.1±10.4 103.0±15.3 103.0±20.9 93.6±29.8
67 55 35 8b
Mean±SEM
2.79±0.1 3.31±0.2 3.75±0.3 3.76±0.5
Friedman test for repeat samples and Wilcoxon signed rank test only to compare LNG values at Days 27–50 with other values. a These figures indicate the number of cases both for age and LNG measurements. b In six samples we were unable to measure LNG due to technical problems.
86
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89
of 20–5000 pg/mL. The run time was 5.5 min and the retention time of the LNG was 3.8 (±0.3) min [18]. E2 levels were measured by electrochemiluminescence immunoassay using a commercial kit (Roche Diagnostics, Mannheim, Germany) with a measuring range of 5.0– 4300 pg/mL and interassay coefficient of variation of 4.9%. All samples were measured in duplicate. 2.2. Statistical analysis LNG levels and endometrial thickness were compared using the Friedman analysis for repeat samples and the Wilcoxon signed rank test or χ2 test when appropriate. Bleeding patterns and endometrial thickness were compared using the Mann–Whitney multiple comparisons after Kruskal–Wallis test. BMI (kg/m2) and weight were compared with LNG levels using the Pearson linear correlation. Significance was established at pb.05. All data are presented as mean±SEM. 3. Results The number of women recruited at the beginning of the study was 86; however, the number of women attending each follow-up visit varied. The mean age±SEM of the 86 women at the beginning of the study was 33.4±0.8 years (range, 20– 47 years); and 23 (26.7%), 49 (57%) and 14 (16.3%) women were in the 20- to 29-year, 30- to 39-year, and 40- to 49-year age bracket, respectively. At 84 months of exposure, 69 women returned to the clinic, this number decreasing over time as many women requested removal of the device and some requested replacement. Two women were excluded because they were postmenopausal at 84 months of evaluation. The mean age of the 67 women remaining in the study at 7 years of use was 34.3±0.8 years (range, 25–49 years). Of these 67 women, 17 (25.4%), 37 (55.2%) and 13 (19.4%) women were in the 20- to 29-year, 30- to 39-year, and 40- to 49-year age bracket, respectively, at 84 months of use. Of these 67 women, three, six, and nine women in the 20to 29-year bracket discontinued at 90, 96 and 102 months, respectively. The discontinuations at the age bracket of 30– 39 years were 9, 11 and 12 women, respectively, and at the age 40–49 years were four, nine and two women, respectively. BMI (kg/m2) mean±SEM at 84 months of evaluation was 26.8±0.67 (range, 18.3–47.3). Table 2 Bleeding patterns at different times after insertion of the LNG-IUS Bleeding pattern
Amenorrhea Spotting Oligomenorrhea Normal menses Total
Fig. 1. Bleeding pattern (%) of patients using the LNG-IUS at four evaluation times.
Mean LNG levels (±SEM) were 253±27 pg/mL (range, 86–760) in the first 2 months following insertion of the device. At the seventh year of use (84 months), LNG levels (±SEM) were 137±12 (range, 23–393) (pb.006; Wilcoxon signed rank test). After that, LNG levels decrease to 119±9 pg/mL (range, 110–129) at 8 1/2 years (102 months) of use although without significance when compared to Month 84 of use. The five LNG values above 300 pg/mL occurred in three women one time and one woman in two determinations. In the aggregate, although all values modified the percentage of change of LNG values between determinations, the 300 pg/mL values do not modify the significance of the results. The corresponding mean endometrial thickness at 84 months (±SEM) of use was 2.79±0.1 mm, increasing to 3.75±0.3 and 3.76±0.5 mm at 96 and 102 months of use, respectively. All comparisons between Month 84 and the others were statistically significant (pb.0001) with the exception of the comparison between the values recorded at the 90th and 96th and 90th, 96th and 102th months (Friedman test for repeat samples) (Table 1). Mean E2 levels (±SEM) ranged from 99.2±11.4 at 27–50 days of use to 86.1±10.4 at 84 months of use and then to 93.6±29.8 pg/mL at 102 months of use. Bleeding patterns are shown in Table 2 and Fig. 1. Amenorrhea decreased from around 42% at the 84th month of use to 37.5% at the 102nd month of use. As the number of cases of amenorrhea diminished, the number of women with oligomenorrhea and normal menses increased proportionally. The serum LNG levels were compared with the bleeding patterns at different months after insertion of the Table 3 Mean serum LNG levels according to bleeding pattern and time after insertion of the LNG-IUS Bleeding pattern
Months after insertion
n
pg/mL
n
pg/mL
n
pg/mL
Amenorrhea Spotting Oligomenorrhea Normal menses Total
28 27 5 7 67
121 146 178 133
17 23 3 7 50
144 120 169 110
8 7 3 6 24
160 139 109 100
Months after insertion 84
90
96
102
n
%
n
%
n
%
n
%
28 27 5 7 67
41.8 40.3 7.5 10.4 100.0
21 23 3 7 54
38.9 42.5 5.6 13.0 100.0
12 12 4 6 34
35.3 35.3 11.8 17.6 100.0
3 2 1 2 8
37.5 25.0 12.5 25.0 100.0
84
90
96
Month 102 was not shown due to the small number of cases. Mann–Whitney multiple comparisons after Kruskal–Wallis test (pb.05) to evaluate LNG serum levels among women with different bleeding patterns.
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89 Table 4 Comparison between LNG and E2 levels and endometrial thickness among women with LNG below 100 pg/mL Month of evaluation
n
LNG (pg/mL)
E2 (pg/mL)
Endometrial thickness (in mm)
84 90 96
22 20 13
68±4 60±5 72±3
116.4±28.1 115.7±31.2 88.0±18.8
2.72±0.2 3.52±0.4 4.13±0.8
87
The linear correlation between women's weight and LNG serum levels at 84th, 90th and 96th months of evaluation showed a p value of .0138, .0004, and .2530, respectively, showing a trend that women with higher weight had lower LNG levels.
4. Discussion
All values are mean±SEM.
device for fewer women than expected due to problems with LNG determinations; however, no correlation was found between the two variables (Table 3). The mean (±SEM) endometrium thickness was 2.37±0.1 and 3.37±0.2 among women with amenorrhea and women who had normal menses (pb.006; Mann–Whitney test). In another analysis performed with women who presented with low LNG values (b100 pg/mL), we correlated the E2 levels and endometrial thickness at Months 84, 90 and 96 of evaluation. This analysis showed that despite the low LNG levels, the E2 levels remained in the range observed at the follicular phase level of the normal menstrual cycle at the three times of evaluation. The mean (±SEM) endometrial thickness increased significantly from 2.7 at 84 months to 4.1 at Month 96 (Table 4). Fig. 2 shows the correlation between serum LNG levels and BMI (kg/m2), and a trend may be observed between high BMI (kg/m2) and low serum LNG levels. Nevertheless, obese women with BMI (kg/m2) ≥35 did not show a correlation with LNG levels at the 84th month; although at 90 months of evaluation, the serum LNG for obese women was 52±10 pg/mL (mean±SEM) and for nonobese women was significantly higher (153±10 pg/mL; p=.0016). At Month 96, the values were not affected by the body weight, and at Month 102, there were no obese women.
The results of our study show that after 7 years of use, LNG levels were almost half of the values observed in the first 2 months after insertion of the LNG-IUS. However, a great variation was observed in serum LNG concentration, some women having lower LNG levels at the beginning of use than others at the eighth year of use. Mean levels of LNG ranged from 760 pg/mL in the first 2 months of use to 18 pg/ mL at the 90th month of use. These variations could have been accounted for by the BMI (kg/m2) of the users; however, the women were in general neither overweight nor obese, with a mean BMI of 26.8 kg/m2 at the seventh year of use. Moreover, the correlation between high BMI (kg/m2) and low serum LNG levels was weak, although a correlation was observed between weight and LNG levels at the 84th and 90th month of use, higher weight correlated with low LNG levels. Since the initial studies with the LNG-IUS, substantial variations in LNG levels have been reported in users of the device at different times after insertion. However, it is important to take into account that at the beginning of the development of this contraceptive method, some of the devices were loaded with greater amounts of LNG than the current device. In addition, these devices may have been releasing LNG at different rates and, moreover, the technique used to measure LNG at that time was radioimmunoassay instead of the technique currently used [19–21].
Fig. 2. Linear correlation between LNG levels and BMI (kg/m2) of women using the LNG-IUS at three evaluation times (Pearson linear correlation).
88
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89
Another explanation for the variations in serum LNG levels may include the levels of sex hormone-binding globulin (SHBG), since an inverse correlation has been found between LNG and SHBG serum levels, and anovulatory women have higher SHBG levels compared to women with ovulatory cycles [1,22]. Unfortunately, progesterone was not measured to evaluate whether the women were ovulatory or anovulatory. Additionally, only 1–2% of the total serum LNG levels represent the free steroid, and serum SHBG levels do not change significantly following insertion of the LNG-IUS or an LNG-releasing vaginal ring [1,22]. The E2 levels found in the first 2 months of use were similar to those found 81/2 years after insertion and are also similar to those found in the follicular phase of the menstrual cycle of normally menstruating women. This finding has been described previously [1,23] and is considered important since it assures both physicians and users that E2 levels remain unaffected by circulating LNG levels and confirms that there was no relationship between bleeding pattern and E2 levels. The present results should not encourage physicians to allow women to retain the same LNG-IUS beyond the approved 5 years of use. According to these data, extended use beyond the approved life span of the device is not recommended. The principal reason for this precaution is the concern that the zero pregnancy rate may result from the fact that the study was not designed to evaluate contraceptive performance and that the sample size was too small to allow any conclusion to be drawn on this subject. Secondly, almost 75% of the women were 30 years of age or older at last follow-up, an age at which fertility begins to decline [24], and even if there was no LNG left in the device, at least these women were using a plastic IUD that provides contraception, albeit without the high efficacy of the LNGIUS [25]. Third, amenorrhea is one of the characteristic effects of the LNG-IUS, and the number of women with amenorrhea diminished inversely as a function of the time of use of the device. A previous study [2] reported that in women who replaced the LNG-IUS with a new device, the proportion of amenorrhea was higher after the second insertion when compared to the first segment of use. However, it is important to state that bleeding patterns were unrelated to contraceptive efficacy [3]. In conclusion, the present data show that after the 7th year of use of the LNG-IUS, LNG levels decline significantly when compared to the initial months of use. This decline in LNG levels provoked a change in the bleeding patterns of women, with amenorrhea becoming less common and women reporting bleeding more frequently compared to their patterns at the beginning of use. The extended use of this device should not be recommended under any circumstances since it is unknown whether circulating LNG levels are sufficient to maintain contraceptive efficacy and also because one of the principal benefits of the device, which is reduced menstrual bleeding
or amenorrhea, is less common. Nevertheless, at the end of the fifth year of use, physicians and users may remain confident that the window of opportunity for replacing the device is wide and should not, therefore, be concerned if the device can only be replaced a few months after its life span. Acknowledgments The authors would like to acknowledge the contribution of Dr. Rafael Barrientos Astigarra of MAGABI Pesquisa Clinica Farmacêutica, São Paulo, Brazil, for performing the LNG measurements and of the Fundação de Amparo a Pesquisa do Estado de São Paulo, Brazil, for providing partial financial support under Award #03/083917. References [1] Luukkainen T, Lähteenmäki P, Toivonen J. Levonorgestrel-releasing intrauterine device. Ann Med 1990;22:85–90. [2] Ronnerdag M, Odlind V. Health effects of long-term use of the intrauterine levonorgestrel-releasing system. A follow-up study over 12 years of continuous use. Acta Obstet Gynecol Scand 1999;78: 716–21. [3] Hidalgo M, Bahamondes L, Perrotti M, Diaz J, Dantas-Monteiro C, Petta C. Bleeding patterns and clinical performance of the levonorgestrel-releasing intrauterine system (Mirena) up to two years. Contraception 2002;65:129–32. [4] Monteiro I, Bahamondes L, Diaz J, Perrotti M, Petta C. Therapeutic use of levonorgestrel-releasing intrauterine system in women with menorrhagia: a pilot study. Contraception 2002;65:325–8. [5] Hurskainen R, Teperi J, Rissanen P, et al. Clinical outcomes and costs with the levonorgestrel-releasing intrauterine system or hysterectomy for treatment of menorrhagia: randomized trial 5-year follow-up. JAMA 2004;291:1456–63. [6] Raudaskoski T, Tapanainen J, Tomas E, et al. Intrauterine 10 μg and 20 μg levonorgestrel systems in postmenopausal women receiving oral oestrogen replacement therapy: clinical, endometrial and metabolic response. BJOG 2002;109:136–44. [7] Hampton NRE, Rees MCP, Lowe DG, Rauramo I, Barlow D, Guillebaud J. Levonorgestrel intrauterine system (LNG-IUS) with conjugated oral equine estrogen: a successful regimen for HRT in perimenopausal women. Hum Reprod 2005;20:2653–60. [8] Petta CA, Ferriani RA, Abrao MS, et al. Randomized clinical trial of a levonorgestrel-releasing intrauterine system and a depot GnRH analogue for the treatment of chronic pelvic pain in women with endometriosis. Hum Reprod 2005;20:1993–8. [9] Lockhat FB, Emembolu JO, Konje JC. The efficacy, side-effects and continuation rates in women with symptomatic endometriosis undergoing treatment with an intra-uterine administered progestogen (levonorgestrel): a 3 year follow-up. Hum Reprod 2005;20: 789–93. [10] Vereide AB, Arnes M, Straume B, Maltau JM, Orbo A. Nuclear morphometric changes and therapy monitoring in patients with endometrial hyperplasia: a study comparing effects of intrauterine levonorgestrel and systemic medroxyprogesterone. Gynecol Oncol 2003;91:526–33. [11] Bahamondes L, Ribeiro-Huguet P, de Andrade KC, Leon-Martins O, Petta CA. Levonorgestrel-releasing intrauterine system (Mirena) as a therapy for endometrial hyperplasia and carcinoma. Acta Obstet Gynecol Scand 2003;82:580–2. [12] World Health Organization. Medical eligibility criteria for contraceptive use. 3rd ed. Geneva: WHO; 2004.
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89 [13] Sivin I, Stern J, Coutinho E, et al. Prolonged intrauterine contraception: a seven-year randomized study of the levonorgestrel 20 mcg/day (LNg 20) and the Copper T380 Ag IUDs. Contraception 1991;44:473–80. [14] Faúndes A, Alvarez F, Díaz J. A Latin American experience with levonorgestrel IUD. Ann Med 1993;25:149–53. [15] Diaz J, Bahamondes L, Monteiro I, Petta C, Hildalgo MM, Arce XE. Acceptability and performance of the levonorgestrel-releasing intrauterine system (Mirena) in Campinas, Brazil. Contraception 2000;62: 59–61. [16] Rodriguez G, Faundes A, Atkinson L. An approach to the analysis of menstrual patterns in the clinical evaluation of contraceptives. Stud Fam Plann 1976;7:42–51. [17] Belsey EM, Machin D, D'Arcangues C. The analysis of vaginal bleeding patterns induced by fertility regulating methods. Contraception 1986;34:253–60. [18] Theron HB, Coetzee C, Sutherland FCW, Wiesner JL, Swart KJ. Selective and sensitive liquid chromatography–tandem mass spectrometry method for the determination of levonorgestrel in human plasma. J Chromatogr B Analyt Technol Biomed Life Sci 2004;813:331–6. [19] Nilsson CG, Lahteenmaki P, Robertson DN, Luukkainen T. Plasma concentrations of levonorgestrel as a function of the release rate of
[20]
[21]
[22]
[23]
[24]
[25]
89
levonorgestrel from medicated intra-uterine devices. Acta Endocrinol (Copenh) 1980;93:380–4. Nilsson CG, Luukkainen T, Lähteenmäki P. Determination of plasma concentrations of D-norgestrel during a one year follow-up in women with a D-norgestrel-releasing IUD. Contraception 1978;17:569–73. Nilsson CG, Lähteenmäki P, Luukkainen T. Levonorgestrel plasma concentrations and hormone profiles after insertion and after one year of treatment with a levonorgestrel-IUD. Contraception 1980;21: 225–33. Jia MC, Zhou LY, Dong L, Xiao B. Serum SHBG levels during normal menstrual cycle and after insertion of levonorgestrel-releasing IUD. Adv Contracept 1992;8:33–40. Nilsson CG, Lahteenmaki PL, Luukkainen T. Ovarian function in amenorrheic and menstruating users of a levonorgestrel-releasing intrauterine device. Fertil Steril 1984;41:52–5. Schwartz D, Mayaux MJ. Female fecundity as a function of age: results of artificial insemination in 2193 nulliparous women with azoospermic husbands. Federation CECOS. N Engl J Med 1982;306:404–6. Díaz J, Díaz MM, Pastene L, Araki R, Faúndes A. Randomized clinical study of the T-Cu 380A and the Lippes loop C, in Campinas, Brazil. Contraception 1982;26:221–8.
Original research article
Serum levonorgestrel levels and endometrial thickness during extended use of the levonorgestrel-releasing intrauterine system Maria M. Hidalgo, Creusa Hidalgo-Regina, M. Valeria Bahamondes, Ilza Monteiro, Carlos A. Petta, Luis Bahamondes⁎ Human Reproduction Unit, Department of Obstetrics and Gynecology, School of Medicine, Universidade Estadual de Campinas (UNICAMP), 13084-971 Campinas, Brazil Received 18 June 2008; revised 8 January 2009; accepted 8 January 2009
Abstract Background: The levonorgestrel-releasing intrauterine system (LNG-IUS) is a contraceptive method approved for five years of use. However, there is some evidence that its life span may be longer. The aim of the study was to evaluate serum levonorgestrel (LNG) and estradiol (E2) levels and endometrial thickness every 6 months from 7 to 8 1/2 years after insertion. Study Design: At the end of the approved 5-year life span, no replacement devices were available; therefore, 86 women were allowed to retain the same device for a further 2 years. At the 7-year follow-up visit, the women who consented were again allowed to retain the same device for a further 18 months and were followed-up at six-monthly intervals. At each visit, vaginal sonography was performed to measure endometrial thickness, and a blood sample was taken to measure LNG and E2. Results: Eighty-four months after insertion, 67 women aged 34.3±0.8 years (mean±SEM) (range, 25–49 years) returned for follow-up. Mean±SEM LNG levels decreased from an initial 253±27 pg/mL (range, 86–760) during the first 2 months following insertion to 137±12 (range, 23–393) at 84 months and 119±9 pg/mL (range, 110–129) at 102 months of use (±SEM). At 84 months of use, mean±SEM endometrial thickness was 2.8±0.1 mm, increasing to 3.8±0.5 mm at 102 months of use. The incidence of amenorrhea decreased from 41.8% at 84 months to 31.5% at 102 months of use. No correlation was found between LNG levels and bleeding patterns; however, a weak correlation was found between high body mass index (kg/m2), high weight, and low serum LNG levels. E2 levels were similar to those of the follicular phase of the menstrual cycle of regularly menstruating women. Conclusions: During extended use of the LNG-IUS, serum LNG levels were nearly half those found in the first 2 months of use (Wilcoxon signed rank test); serum E2 levels were normal. Despite the very thin endometrium, menstrual bleeding was reinstated in many cases. At the end of its 5-year life span, there is a window for changing the LNG-IUS, and physicians and users should not be concerned about delaying replacement of the device for a short time beyond the approved life span; however, maintaining the same device long after its approved life span cannot be recommended. © 2009 Elsevier Inc. All rights reserved. Keywords: Levonorgestrel-releasing intrauterine system; Mirena®; Levonorgestrel; Contraception
1. Introduction Long-term contraceptive methods are desired by many women because they provide contraception without the requirement of having to remember to take a daily pill or a monthly or three-monthly injection. Long-term methods include subdermal implants, intrauterine devices and the levonorgestrel-releasing intrauterine system (LNG-IUS). ⁎ Corresponding author. Tel.: +55 19 3289 2856; fax: +55 19 3289 2440. E-mail address: [email protected] (L. Bahamondes). 0010-7824/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2009.01.004
The LNG-IUS is marketed and approved in 123 countries for use as a contraceptive method. One of the effects of the use of the LNG-IUS is amenorrhea caused by the antiproliferative effect of levonorgestrel (LNG) on the endometrium [1–3]. As a consequence of this effect, the device has been approved and is currently in use in many countries as a treatment for idiopathic menorrhagia [4,5] and as endometrial protection during estrogen therapy [6,7]. In addition, this device has been used off-label as a treatment for endometriosis [8,9] and for endometrial hyperplasia [10,11].
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89
The LNG-IUS is approved for 5 years of use during which time the device provides high contraceptive efficacy with few adverse effects [1,12]. Some limited data exist on the device's extended use beyond 5 years and up to 7 years [2,13,14]. The objectives of this study were to evaluate the serum levels of LNG and estradiol (E2) and the endometrial thickness of women who used the LNG-IUS beyond its approved life span of 5 years.
2. Materials and methods The study was conducted at the Human Reproduction Unit, Department of Obstetrics and Gynecology, School of Medical Sciences, Universidade Estadual de Campinas, Campinas, Brazil. The ethical committee of the university approved the study, and all participants signed an informed consent form prior to admission. The study population consisted of a subsample from a larger group of women participating in an acceptability study of the LNG-IUS (Mirena®, Bayer Schering Pharma Oy, Turku, Finland) in Brazil who had their devices inserted between April and September 1998 [3,15]. At the end of the approved 5-year life span of the device, no new devices were available to replace the ones in use; therefore, any woman who wished to do so were allowed to retain the same device for a further 2 years. At the 7-year follow-up visit, the women in the extended use protocol were once again allowed to retain the same device for a further 18 months and were invited to return for follow-up visits at six-monthly intervals. At each visit, the threads of the LNG-IUS were visualized during a gynecological examination. Next, ultrasonography was performed using a 6.0-MHz vaginal probe (Justavision 400, Toshiba, Toshigi-Ken, Japan) to measure double-layer endometrial thickness and to be sure that the LNG-IUS was located in the uterine fundus. A blood sample was then collected, and the serum was separated and stored at −20°C until assayed. Weight for each woman was evaluated at each follow-up visit and body mass index (BMI, kg/m2) was calculated. Also, we reviewed and retrieved the
85
data from the menstrual calendar kept by each woman during the 90 days preceding the interview. According to the World Health Organization [16,17], bleeding was defined as “any bloody vaginal discharge that requires the use of protection such as pads and tampons” and spotting as “any bloody vaginal discharge that is not intense enough to require sanitary protection”. Amenorrhea was defined as no bleeding during the preceding 90 days before the interview and oligomenorrhea as bleeding episodes with more than 35 days interval. Serum samples from women who had insertion of an LNG-IUS in 1998 had been collected at the first scheduled follow-up visit after insertion, and aliquots were kept in the freezer. In this clinic, the first follow-up visit was generally scheduled 30–50 days after insertion. Blood was collected 27–50 days following insertion. Follow-up visits were carried out at six-monthly intervals from 84 months until 102 months after insertion. 2.1. LNG and E2 assay Serum LNG levels were measured using a validated method of high-performance liquid chromatography coupled with tandem mass spectrometry. All procedures were carried out in compliance with the guidelines approved by the Brazilian regulatory authorities (ANVISA) for Good Laboratory Practices. The LNG standards used in the assay were obtained from USP (Rockville, MD, USA). Biological specificity of the method was checked by processing independent serum samples and blank samples obtained from women not using any kind of hormones. Briefly, the bioanalytical assay for the quantification of LNG was developed using an online solid phase extraction method (Symbiosis, Spark Holland, Emmen, The Netherlands). The mass spectrometer (Sciex/Applied Biosystems, model API5000, Toronto, Canada) was equipped with a photo-ionization source (APPI) running in positive ion mode setup in Multiple Reaction Monitoring for the transition m/z 313.3 245.1 for LNG. Toluene was used as solvent at the flow rate of 0.15 mL/min. The lowest limit of quantification of the method was 20 pg/mL, and it was linear over the range
Table 1 Serum LNG levels (pg/mL) and endometrial thickness (mm) at different times after insertion of the LNG-IUS Time after insertion
Days 27–50 Months 84 90 96 102
Age at blood collection
LNG (pg/mL)
Mean±SEM
na
Mean±SEM
33.4±0.71
86
253±27
86–760
34.3±0.75 35.2±1.00 34.4±1.20 41.5±3.50
67 51 25 2b
137±12 130±11 132±18 119±9
23–393 18–470 51–505 110–129
Range
Endometrial thickness (mm)
E2 (pg/mL)
n
n
Mean±SEM
86
99.2±11.4
67 58 34 8b
86.1±10.4 103.0±15.3 103.0±20.9 93.6±29.8
67 55 35 8b
Mean±SEM
2.79±0.1 3.31±0.2 3.75±0.3 3.76±0.5
Friedman test for repeat samples and Wilcoxon signed rank test only to compare LNG values at Days 27–50 with other values. a These figures indicate the number of cases both for age and LNG measurements. b In six samples we were unable to measure LNG due to technical problems.
86
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89
of 20–5000 pg/mL. The run time was 5.5 min and the retention time of the LNG was 3.8 (±0.3) min [18]. E2 levels were measured by electrochemiluminescence immunoassay using a commercial kit (Roche Diagnostics, Mannheim, Germany) with a measuring range of 5.0– 4300 pg/mL and interassay coefficient of variation of 4.9%. All samples were measured in duplicate. 2.2. Statistical analysis LNG levels and endometrial thickness were compared using the Friedman analysis for repeat samples and the Wilcoxon signed rank test or χ2 test when appropriate. Bleeding patterns and endometrial thickness were compared using the Mann–Whitney multiple comparisons after Kruskal–Wallis test. BMI (kg/m2) and weight were compared with LNG levels using the Pearson linear correlation. Significance was established at pb.05. All data are presented as mean±SEM. 3. Results The number of women recruited at the beginning of the study was 86; however, the number of women attending each follow-up visit varied. The mean age±SEM of the 86 women at the beginning of the study was 33.4±0.8 years (range, 20– 47 years); and 23 (26.7%), 49 (57%) and 14 (16.3%) women were in the 20- to 29-year, 30- to 39-year, and 40- to 49-year age bracket, respectively. At 84 months of exposure, 69 women returned to the clinic, this number decreasing over time as many women requested removal of the device and some requested replacement. Two women were excluded because they were postmenopausal at 84 months of evaluation. The mean age of the 67 women remaining in the study at 7 years of use was 34.3±0.8 years (range, 25–49 years). Of these 67 women, 17 (25.4%), 37 (55.2%) and 13 (19.4%) women were in the 20- to 29-year, 30- to 39-year, and 40- to 49-year age bracket, respectively, at 84 months of use. Of these 67 women, three, six, and nine women in the 20to 29-year bracket discontinued at 90, 96 and 102 months, respectively. The discontinuations at the age bracket of 30– 39 years were 9, 11 and 12 women, respectively, and at the age 40–49 years were four, nine and two women, respectively. BMI (kg/m2) mean±SEM at 84 months of evaluation was 26.8±0.67 (range, 18.3–47.3). Table 2 Bleeding patterns at different times after insertion of the LNG-IUS Bleeding pattern
Amenorrhea Spotting Oligomenorrhea Normal menses Total
Fig. 1. Bleeding pattern (%) of patients using the LNG-IUS at four evaluation times.
Mean LNG levels (±SEM) were 253±27 pg/mL (range, 86–760) in the first 2 months following insertion of the device. At the seventh year of use (84 months), LNG levels (±SEM) were 137±12 (range, 23–393) (pb.006; Wilcoxon signed rank test). After that, LNG levels decrease to 119±9 pg/mL (range, 110–129) at 8 1/2 years (102 months) of use although without significance when compared to Month 84 of use. The five LNG values above 300 pg/mL occurred in three women one time and one woman in two determinations. In the aggregate, although all values modified the percentage of change of LNG values between determinations, the 300 pg/mL values do not modify the significance of the results. The corresponding mean endometrial thickness at 84 months (±SEM) of use was 2.79±0.1 mm, increasing to 3.75±0.3 and 3.76±0.5 mm at 96 and 102 months of use, respectively. All comparisons between Month 84 and the others were statistically significant (pb.0001) with the exception of the comparison between the values recorded at the 90th and 96th and 90th, 96th and 102th months (Friedman test for repeat samples) (Table 1). Mean E2 levels (±SEM) ranged from 99.2±11.4 at 27–50 days of use to 86.1±10.4 at 84 months of use and then to 93.6±29.8 pg/mL at 102 months of use. Bleeding patterns are shown in Table 2 and Fig. 1. Amenorrhea decreased from around 42% at the 84th month of use to 37.5% at the 102nd month of use. As the number of cases of amenorrhea diminished, the number of women with oligomenorrhea and normal menses increased proportionally. The serum LNG levels were compared with the bleeding patterns at different months after insertion of the Table 3 Mean serum LNG levels according to bleeding pattern and time after insertion of the LNG-IUS Bleeding pattern
Months after insertion
n
pg/mL
n
pg/mL
n
pg/mL
Amenorrhea Spotting Oligomenorrhea Normal menses Total
28 27 5 7 67
121 146 178 133
17 23 3 7 50
144 120 169 110
8 7 3 6 24
160 139 109 100
Months after insertion 84
90
96
102
n
%
n
%
n
%
n
%
28 27 5 7 67
41.8 40.3 7.5 10.4 100.0
21 23 3 7 54
38.9 42.5 5.6 13.0 100.0
12 12 4 6 34
35.3 35.3 11.8 17.6 100.0
3 2 1 2 8
37.5 25.0 12.5 25.0 100.0
84
90
96
Month 102 was not shown due to the small number of cases. Mann–Whitney multiple comparisons after Kruskal–Wallis test (pb.05) to evaluate LNG serum levels among women with different bleeding patterns.
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89 Table 4 Comparison between LNG and E2 levels and endometrial thickness among women with LNG below 100 pg/mL Month of evaluation
n
LNG (pg/mL)
E2 (pg/mL)
Endometrial thickness (in mm)
84 90 96
22 20 13
68±4 60±5 72±3
116.4±28.1 115.7±31.2 88.0±18.8
2.72±0.2 3.52±0.4 4.13±0.8
87
The linear correlation between women's weight and LNG serum levels at 84th, 90th and 96th months of evaluation showed a p value of .0138, .0004, and .2530, respectively, showing a trend that women with higher weight had lower LNG levels.
4. Discussion
All values are mean±SEM.
device for fewer women than expected due to problems with LNG determinations; however, no correlation was found between the two variables (Table 3). The mean (±SEM) endometrium thickness was 2.37±0.1 and 3.37±0.2 among women with amenorrhea and women who had normal menses (pb.006; Mann–Whitney test). In another analysis performed with women who presented with low LNG values (b100 pg/mL), we correlated the E2 levels and endometrial thickness at Months 84, 90 and 96 of evaluation. This analysis showed that despite the low LNG levels, the E2 levels remained in the range observed at the follicular phase level of the normal menstrual cycle at the three times of evaluation. The mean (±SEM) endometrial thickness increased significantly from 2.7 at 84 months to 4.1 at Month 96 (Table 4). Fig. 2 shows the correlation between serum LNG levels and BMI (kg/m2), and a trend may be observed between high BMI (kg/m2) and low serum LNG levels. Nevertheless, obese women with BMI (kg/m2) ≥35 did not show a correlation with LNG levels at the 84th month; although at 90 months of evaluation, the serum LNG for obese women was 52±10 pg/mL (mean±SEM) and for nonobese women was significantly higher (153±10 pg/mL; p=.0016). At Month 96, the values were not affected by the body weight, and at Month 102, there were no obese women.
The results of our study show that after 7 years of use, LNG levels were almost half of the values observed in the first 2 months after insertion of the LNG-IUS. However, a great variation was observed in serum LNG concentration, some women having lower LNG levels at the beginning of use than others at the eighth year of use. Mean levels of LNG ranged from 760 pg/mL in the first 2 months of use to 18 pg/ mL at the 90th month of use. These variations could have been accounted for by the BMI (kg/m2) of the users; however, the women were in general neither overweight nor obese, with a mean BMI of 26.8 kg/m2 at the seventh year of use. Moreover, the correlation between high BMI (kg/m2) and low serum LNG levels was weak, although a correlation was observed between weight and LNG levels at the 84th and 90th month of use, higher weight correlated with low LNG levels. Since the initial studies with the LNG-IUS, substantial variations in LNG levels have been reported in users of the device at different times after insertion. However, it is important to take into account that at the beginning of the development of this contraceptive method, some of the devices were loaded with greater amounts of LNG than the current device. In addition, these devices may have been releasing LNG at different rates and, moreover, the technique used to measure LNG at that time was radioimmunoassay instead of the technique currently used [19–21].
Fig. 2. Linear correlation between LNG levels and BMI (kg/m2) of women using the LNG-IUS at three evaluation times (Pearson linear correlation).
88
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89
Another explanation for the variations in serum LNG levels may include the levels of sex hormone-binding globulin (SHBG), since an inverse correlation has been found between LNG and SHBG serum levels, and anovulatory women have higher SHBG levels compared to women with ovulatory cycles [1,22]. Unfortunately, progesterone was not measured to evaluate whether the women were ovulatory or anovulatory. Additionally, only 1–2% of the total serum LNG levels represent the free steroid, and serum SHBG levels do not change significantly following insertion of the LNG-IUS or an LNG-releasing vaginal ring [1,22]. The E2 levels found in the first 2 months of use were similar to those found 81/2 years after insertion and are also similar to those found in the follicular phase of the menstrual cycle of normally menstruating women. This finding has been described previously [1,23] and is considered important since it assures both physicians and users that E2 levels remain unaffected by circulating LNG levels and confirms that there was no relationship between bleeding pattern and E2 levels. The present results should not encourage physicians to allow women to retain the same LNG-IUS beyond the approved 5 years of use. According to these data, extended use beyond the approved life span of the device is not recommended. The principal reason for this precaution is the concern that the zero pregnancy rate may result from the fact that the study was not designed to evaluate contraceptive performance and that the sample size was too small to allow any conclusion to be drawn on this subject. Secondly, almost 75% of the women were 30 years of age or older at last follow-up, an age at which fertility begins to decline [24], and even if there was no LNG left in the device, at least these women were using a plastic IUD that provides contraception, albeit without the high efficacy of the LNGIUS [25]. Third, amenorrhea is one of the characteristic effects of the LNG-IUS, and the number of women with amenorrhea diminished inversely as a function of the time of use of the device. A previous study [2] reported that in women who replaced the LNG-IUS with a new device, the proportion of amenorrhea was higher after the second insertion when compared to the first segment of use. However, it is important to state that bleeding patterns were unrelated to contraceptive efficacy [3]. In conclusion, the present data show that after the 7th year of use of the LNG-IUS, LNG levels decline significantly when compared to the initial months of use. This decline in LNG levels provoked a change in the bleeding patterns of women, with amenorrhea becoming less common and women reporting bleeding more frequently compared to their patterns at the beginning of use. The extended use of this device should not be recommended under any circumstances since it is unknown whether circulating LNG levels are sufficient to maintain contraceptive efficacy and also because one of the principal benefits of the device, which is reduced menstrual bleeding
or amenorrhea, is less common. Nevertheless, at the end of the fifth year of use, physicians and users may remain confident that the window of opportunity for replacing the device is wide and should not, therefore, be concerned if the device can only be replaced a few months after its life span. Acknowledgments The authors would like to acknowledge the contribution of Dr. Rafael Barrientos Astigarra of MAGABI Pesquisa Clinica Farmacêutica, São Paulo, Brazil, for performing the LNG measurements and of the Fundação de Amparo a Pesquisa do Estado de São Paulo, Brazil, for providing partial financial support under Award #03/083917. References [1] Luukkainen T, Lähteenmäki P, Toivonen J. Levonorgestrel-releasing intrauterine device. Ann Med 1990;22:85–90. [2] Ronnerdag M, Odlind V. Health effects of long-term use of the intrauterine levonorgestrel-releasing system. A follow-up study over 12 years of continuous use. Acta Obstet Gynecol Scand 1999;78: 716–21. [3] Hidalgo M, Bahamondes L, Perrotti M, Diaz J, Dantas-Monteiro C, Petta C. Bleeding patterns and clinical performance of the levonorgestrel-releasing intrauterine system (Mirena) up to two years. Contraception 2002;65:129–32. [4] Monteiro I, Bahamondes L, Diaz J, Perrotti M, Petta C. Therapeutic use of levonorgestrel-releasing intrauterine system in women with menorrhagia: a pilot study. Contraception 2002;65:325–8. [5] Hurskainen R, Teperi J, Rissanen P, et al. Clinical outcomes and costs with the levonorgestrel-releasing intrauterine system or hysterectomy for treatment of menorrhagia: randomized trial 5-year follow-up. JAMA 2004;291:1456–63. [6] Raudaskoski T, Tapanainen J, Tomas E, et al. Intrauterine 10 μg and 20 μg levonorgestrel systems in postmenopausal women receiving oral oestrogen replacement therapy: clinical, endometrial and metabolic response. BJOG 2002;109:136–44. [7] Hampton NRE, Rees MCP, Lowe DG, Rauramo I, Barlow D, Guillebaud J. Levonorgestrel intrauterine system (LNG-IUS) with conjugated oral equine estrogen: a successful regimen for HRT in perimenopausal women. Hum Reprod 2005;20:2653–60. [8] Petta CA, Ferriani RA, Abrao MS, et al. Randomized clinical trial of a levonorgestrel-releasing intrauterine system and a depot GnRH analogue for the treatment of chronic pelvic pain in women with endometriosis. Hum Reprod 2005;20:1993–8. [9] Lockhat FB, Emembolu JO, Konje JC. The efficacy, side-effects and continuation rates in women with symptomatic endometriosis undergoing treatment with an intra-uterine administered progestogen (levonorgestrel): a 3 year follow-up. Hum Reprod 2005;20: 789–93. [10] Vereide AB, Arnes M, Straume B, Maltau JM, Orbo A. Nuclear morphometric changes and therapy monitoring in patients with endometrial hyperplasia: a study comparing effects of intrauterine levonorgestrel and systemic medroxyprogesterone. Gynecol Oncol 2003;91:526–33. [11] Bahamondes L, Ribeiro-Huguet P, de Andrade KC, Leon-Martins O, Petta CA. Levonorgestrel-releasing intrauterine system (Mirena) as a therapy for endometrial hyperplasia and carcinoma. Acta Obstet Gynecol Scand 2003;82:580–2. [12] World Health Organization. Medical eligibility criteria for contraceptive use. 3rd ed. Geneva: WHO; 2004.
M.M. Hidalgo et al. / Contraception 80 (2009) 84–89 [13] Sivin I, Stern J, Coutinho E, et al. Prolonged intrauterine contraception: a seven-year randomized study of the levonorgestrel 20 mcg/day (LNg 20) and the Copper T380 Ag IUDs. Contraception 1991;44:473–80. [14] Faúndes A, Alvarez F, Díaz J. A Latin American experience with levonorgestrel IUD. Ann Med 1993;25:149–53. [15] Diaz J, Bahamondes L, Monteiro I, Petta C, Hildalgo MM, Arce XE. Acceptability and performance of the levonorgestrel-releasing intrauterine system (Mirena) in Campinas, Brazil. Contraception 2000;62: 59–61. [16] Rodriguez G, Faundes A, Atkinson L. An approach to the analysis of menstrual patterns in the clinical evaluation of contraceptives. Stud Fam Plann 1976;7:42–51. [17] Belsey EM, Machin D, D'Arcangues C. The analysis of vaginal bleeding patterns induced by fertility regulating methods. Contraception 1986;34:253–60. [18] Theron HB, Coetzee C, Sutherland FCW, Wiesner JL, Swart KJ. Selective and sensitive liquid chromatography–tandem mass spectrometry method for the determination of levonorgestrel in human plasma. J Chromatogr B Analyt Technol Biomed Life Sci 2004;813:331–6. [19] Nilsson CG, Lahteenmaki P, Robertson DN, Luukkainen T. Plasma concentrations of levonorgestrel as a function of the release rate of
[20]
[21]
[22]
[23]
[24]
[25]
89
levonorgestrel from medicated intra-uterine devices. Acta Endocrinol (Copenh) 1980;93:380–4. Nilsson CG, Luukkainen T, Lähteenmäki P. Determination of plasma concentrations of D-norgestrel during a one year follow-up in women with a D-norgestrel-releasing IUD. Contraception 1978;17:569–73. Nilsson CG, Lähteenmäki P, Luukkainen T. Levonorgestrel plasma concentrations and hormone profiles after insertion and after one year of treatment with a levonorgestrel-IUD. Contraception 1980;21: 225–33. Jia MC, Zhou LY, Dong L, Xiao B. Serum SHBG levels during normal menstrual cycle and after insertion of levonorgestrel-releasing IUD. Adv Contracept 1992;8:33–40. Nilsson CG, Lahteenmaki PL, Luukkainen T. Ovarian function in amenorrheic and menstruating users of a levonorgestrel-releasing intrauterine device. Fertil Steril 1984;41:52–5. Schwartz D, Mayaux MJ. Female fecundity as a function of age: results of artificial insemination in 2193 nulliparous women with azoospermic husbands. Federation CECOS. N Engl J Med 1982;306:404–6. Díaz J, Díaz MM, Pastene L, Araki R, Faúndes A. Randomized clinical study of the T-Cu 380A and the Lippes loop C, in Campinas, Brazil. Contraception 1982;26:221–8.