REPRODUCTIVE ENDOCRINOLOGY - PDF Free Download

FERTILITY AND STERILITY VOL. 79, NO. 4, APRIL 2003 Copyright 2003 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. REPRODUCTIVE ENDOCRINOLOGY Therapeutic effects of the levonorgestrelreleasing intrauterine system in the treatment of idiopathic menorrhagia Bilian Xiao, M.D., a Shang-Chun Wu, M.S., a Jie Chong, a Tao Zeng, a Li-Hui Han, M.S., b and Tapani Luukkainen, M.D., Ph.D. c National Research Institute for Family Planning, and Beijing Gynecology and Obstetrics Hospital, Beijing, People s Republic of China Received April 15, 2002; revised and accepted August 7, 2002. Supported by Leiras Oy Pharmaceutical Co., Turku, Finland, grant no. 102-95532. Reprint requests: Bilian Xiao, M.D., National Research Institute for Family Planning, No. 12 Da Hui Si, Haidian District, Beijing 100081, People s Republic of China (FAX: 86-19-6217-3457; E-mail: xiaobl@public.bta.net.cn). a National Research Institute for Family Planning. b Beijing Gynecology and Obstetrics Hospital. c Institute of Biomedicine, University of Helsinki, Helsinki, Finland. 0015-0282/03/$30.00 doi:10.1016/s0015-0282(02) 04913-0 Objective: To investigate the effect of the levonorgestrel-releasing intrauterine system (LNG-IUS) in the treatment of idiopathic menorrhagia. Design: Measurements of menstrual blood loss (MBL), hemoglobin, and serum ferritin before and after LNG-IUS insertion. Setting: National Research Institute for Family Planning and Beijing Gynecology and Obstetrics Hospital, Beijing, People s Republic of China. Patient(s): Thirty-four patients with MBL over 80 ml. Intervention(s): Insertion of the LNG-IUS on cycle days 5 7 and follow-up at 3-month intervals for 3 years. Main Outcome Measure(s): Measurement of MBL, serum ferritin, and hemoglobin for evaluation of efficacy of treatment. Result(s): A significant reduction of MBL to 23.4 ml (78.7% decrease), 26.4 ml (83.8% decrease), 2.7 ml (97.7% decrease), and 13.7 ml (85.0% decrease) at 6, 12, 24, and 36 months, respectively. After 6 months, one-third of the patients experienced amenorrhea, and one-fourth, spotting. Hemoglobin increased significantly from 121.5 g/l preinsertion to 135.5 g/l after 36 months, while serum ferritin levels increased significantly from 21.9 ng/ml before insertion to 92.8 ng/ml after 36 months. In women using the LNG-IUS for 3 4 years, the E 2 levels in 20 samples were 239.4 pmol/l, P levels were 11.1 nmol/l, and serum LNG levels were maintained at an average of 511 pmol/l. Conclusion(s): The significant reduction of MBL and the increase in hemoglobin and serum ferritin levels in the treatment of menorrhagia with the LNG-IUS has great implications for women s reproductive health, particularly in developing countries. (Fertil Steril 2003;79:963 9. 2003 by American Society for Reproductive Medicine.) Key Words: Levonorgestrel-releasing intrauterine system, idiopathic menorrhagia, menstrual blood loss, serum ferritin, hemoglobin The levonorgestrel-releasing intrauterine system (LNG-IUS) is known worldwide for its high efficacy in contraception that lasts for at least 5 years. The effect of a reduction in menstrual blood loss (MBL) as early as 3 months after insertion of the LNG-IUS is the most significant advance over copper-bearing intrauterine devices (IUDs), which increase MBL in the first 2 years. The LNG-IUS was first introduced in China by Luukkainen in 1986 at a national conference on family planning (1). Since then, it has been studied as a contraceptive for more than 10 years (2 4), and it has been accepted as a good IUD, particularly for those who have heavy menstrual bleeding. Idiopathic menorrhagia is one of the common gynecological diseases seen in the outpatient clinics of many general hospitals and gynecology clinics. Women suffering from heavy monthly bleeding are often anemic and in ill health, both physically and mentally. Each month before the expected days of menstruation they are under mental stress at home and at work. Patients in China with menorrhagia often resort to treatment with traditional herbal medicine first and then use steroids, such as norethisterone, megestrol acetate, or even testosterone propionate injectables during heavy bleeding. When medical treatment fails, the last resort is surgical intervention. Curettage is 963

often used as an emergency measure to treat heavy bleeding; endometrial ablation is not very popular and is only performed in a few hospitals. Only when all methods fail after many years of treatment and the patient is very ill does the patient resort to hysterectomy. The fact that use of the LNG-IUS can effectively treat menorrhagia (5, 6) brings hope to these women. The therapeutic effects of the LNG-IUS reported by Luukkainen et al. (5) and by Anderson and Rybo (6), as well as our own experience with the use of the LNG-IUS for contraception for more than 10 years in China, prompted us to investigate the treatment of idiopathic menorrhagia with the system. Its effectiveness in reducing bleeding is mainly due to its local action on the endometrium. This clinical trial of the LNG-IUS for treatment of idiopathic menorrhagia, supported by Leiras Oy, Turku, Finland, was initiated in Beijing in 1996 with the objective of evaluating its therapeutic efficacy in the long-term treatment of menorrhagia and its acceptability among Chinese women. MATERIALS AND METHODS Thirty-four parous women aged 27 43 years who experienced regular and heavy menstrual bleeding with a normal sized or slightly enlarged uterus were recruited into the study. Subjects with known or suspected pregnancy, deformity of the uterine cavity or cervical canal, undiagnosed abnormal bleeding, or any other systemic diseases were excluded. Insignificant or small fibroids, detected by ultrasound, were acceptable. According to the inclusion and exclusion criteria, a total of 34 patients were recruited and they were followed up on a 3-month basis for 3 years. Before the insertion of the LNG-IUS, MBL was measured by Hallberg s alkaline hematin method (7) in two menstrual cycles, and only those with MBL over 80 ml on average were admitted to the study. The measurement of MBL was repeated at months 6, 12, 24, and 36 after insertion. The LNG-IUS is manufactured by Leiras Oy in Finland. It consists of a T-shaped polyethylene frame and an LNGcontaining cylinder covered with a membrane regulating the release of the hormone. The total amount of LNG in the cylinder is 52 mg, and its initial release rate is 20 g per 24 hours. The system releases the active ingredient LNG for up to 5 years at a virtually constant rate. It has been studied as a contraceptive device in China for more than 10 years. The IUS with its inserter was provided in a sterile package. Insertion of the LNG-IUS was performed on day 5 of the menstrual cycle under aseptic conditions. Serum ferritin concentrations were measured by radioimmunoassay (RIA). Hemoglobin levels were measured by the ferricyanate method. Both were measured before insertion and once every 3 months after insertion in the first year and then at yearly follow-up visits. The following analyses are based on samples taken at different time intervals. In some cases, because of very scanty spotting, MBL was not measurable, and in cases of amenorrhea no samples were collected. Therefore, the samples analyzed were fewer than the number of patients at each follow-up visit. Serum ferritin and hemoglobin levels were measured at 3, 6, 12, 24, and 36 months. Serum E 2, P, and LNG concentrations were measured by RIA in 20 samples taken from women who used the LNG- IUS for more than 36 months. RIAs of serum E 2, P, and LNG were performed according to the methods of the World Health Organization Collaborating Center for Research and Reference Services in the Immunoassay of Hormones in Human Reproduction (8, 9). The matched reagents of antisera, standards, and tritiated tracers for each assay were purchased from the World Health Organization Collaborating Center for Research. The withinbatch coefficient of variation (CV) for E 2 was 9.7%; for P, 9.8%; and for LNG, 6.4%. The 20 samples were measured in one batch. Serum ferritin concentrations were measured by RIA using ( 125 I)-ferritin. The RIA kit (IMK 410 with approval number 95-weiyao R-2) was purchased from the China Scientific Research Institute of Atomic Energy in Beijing. The RIA kit consisted of standards, antiserum, 125 I-labeled ferritin tracer, and a laboratory manual. The within-batch CV was 8.8%, and the between-batch CV was 16.8%. Changes in hemoglobin and serum ferritin levels before and after insertion of the LNG-IUS were analyzed by the paired t-test. In analyzing MBL data, Wilcoxon s signed rank test (nonparametric) was used (10). The statistical software used in all analyses was SAS, version 6.12 (SAS Institute, Chapel Hill, NC). The study protocol was reviewed by the Scientific Committee and the Ethics Committee of the National Research Institute for Family Planning. Both Committees acknowledged that use of the LNG-IUS would have a beneficial therapeutic effect on Chinese women suffering from heavy menstrual bleeding, and they approved the suggestion that the clinical trial should be conducted by a research team headed by B.X. RESULTS Thirty-four women were recruited after screening MBL among 49 subjects. All patients recruited had a history of idiopathic menorrhagia and failure in previous treatment with hormones or traditional medicine. The LNG-IUSs were inserted on days 5 7 of the menstrual cycle. A total of 1,125 postinsertion woman-months were assessed. The outcome is shown diagrammatically in Figure 1. Patient Characteristics The mean age of the 34 patients at admission was 35 years (SD 4.4; range, 27 43 years). The mean weight was 964 Xiao et al. Treatment of menorrhagia with LNG-IUS Vol. 79, No. 4, April 2003

FIGURE 1 Trial profile. The total number of women at admission was 34. The total number of months of follow-up was 1,125. Two patients were lost to follow-up (one at 3 months and one at 19 months). Four women expelled the LNG-IUS (at day 2 and at month 1, 3, and 15). Two women underwent premature removal (at month 7 and 16). At the end of the study (36 months), there were 24 women. Two women participated less than 36 months but still had follow-up at 24 and 30 months, respectively. IUSs in four women were removed at the end of the study (at 36, 44, 48, and 52 months). Ten subjects were followed-up more than 48 months. month and the other on the second day after insertion because of heavy bleeding. Two partial expulsions were observed during pelvic examination; the lower end of the vertical stem of the LNG-IUS protruded from the external cervical os at the 6-month and 15-month checkup. The IUS was removed by the attending doctor. Two women were lost to follow-up: one at the third month because the subject moved to another city, and another at 19 months. Two subjects requested removal prematurely, one at 7 months because of prolonged spotting and fatigue and the other at 16 months because of persistent amenorrhea for 7 months. Side Effects Complaints after LNG-IUS insertion were rare except for some abdominal pain (n 5), backache (n 2), headache (n 2), breast tenderness (n 2), acne (n 2), dysmenorrhea (n 2), weight gain over 5 kg (n 6), nausea (n 1), and fatigue (n 1). One subject experienced severe abdominal pain on the second day after insertion when the IUS was expelled. These side effects were well tolerated and did not cause any premature removal of the IUS. In five cases, small fibroids were detected before insertion by ultrasound, but they showed no changes in size at the follow-up visits. Prolonged spotting and amenorrhea were the main complaints. Over one-third of the patients experienced amenorrhea, and one-fourth had scanty spotting. Spotting and amenorrhea often occurred alternately, which means that the amenorrhea was not constant, even though there was a tendency toward an increased incidence of amenorrhea with increased duration of use. These changes in menstrual bleeding were well accepted by the women in comparison with the monthly heavy bleeding they suffered before treatment. 59.9 kg (SD 8.1; range 40 80 kg). The mean height was 162 cm (SD 5.5; range 151 170 cm). The mean body mass index (BMI) was 21.7 (SD 1.9; range 14.3 27.6). Pelvic examination revealed a normal uterus in most cases. In five cases, small subserous or intramural fibroids were found by ultrasound. No changes in the size of fibroids were noticed at follow-up examinations. Pregnancy No intrauterine or ectopic pregnancies occurred during the follow-up period of 1,125 woman-months. Expulsion, Loss to Follow-up, and Premature Removal There were four expulsions of the LNG-IUS in total. There were two complete expulsions: one during the first Changes in MBL After insertion of the IUS, MBL became markedly reduced. There were many cycles where MBL was so scanty that it was not measurable, and some patients were amenorrheic. After 6 months, about one-third of the patients experienced amenorrhea and one-fourth, scanty spotting. In Table 1, the amounts of MBL before and after LNG- IUS insertion are given. When compared with the average MBL before insertion (124.2 ml), MBLs at 6, 12, 24, and 36 months were 23.4, 26.4, 2.7, and 13.7 ml, representing reductions of 78.7%, 83.8%, 97.7%, and 85.0%, respectively. On average, the reduction of MBL was 86.3%. Changes of MBL before and after LNG-IUS insertion in each individual are shown graphically in Figure 2. Changes in Serum Ferritin and Hemoglobin Levels Along with the reduction of MBL, there were also significant increases of hemoglobin and serum ferritin. Hemoglobin increased from 121.5 g/l preinsertion to 135.5 g/l after 36 months; serum ferritin increased from 21.9 ng/ml FERTILITY & STERILITY 965

TABLE 1 Menstrual blood loss (MBL) before and after insertion of the levonorgestrel-releasing system. Preinsertion mean of two cycles Postinsertion 6 months 12 months 24 months 36 months No. of subjects 34 29 29 26 23 No. of measurable samples 34 16 13 8 2 Mean (ml) 124.2 23.4 26.4 2.7 13.7 SD 42.9 33.6 42.6 27.1 3.3 Minimum 80.0 0.5 1.6 0 10.3 Maximum 277.5 165.5 133.0 10.5 17.1 P.0001.0001.0001.0001 Amenorrhea 5 6 11 5 Spotting 8 10 7 16 % of reduction of MBL 78.0 83.8 97.7 85.0 Note: In analyzing the MBL data, a nonparametric Wilcoxon signed rank test was used. In these analyses, for patients with amenorrhea or spotting, values 0 and 0.5 were used, respectively. The P-values at each segment month are values for the comparison with the preinsertion data. The number of subjects with amenorrhea or spotting at each segment month and the amount of reduction of MBL are tabulated in the last three rows. before insertion to 92.8 ng/ml at 36 months. The differences were highly significant (Tables 2 and 3). There was one exceptional patient whose preinsertion MBLs were 259.6 and 150 ml. Her MBL was reduced to 100 ml at 12 months, remaining higher than the limit of menorrhagia (80 ml). However, at the 24-month follow-up she was amenorrheic. Her hemoglobin level increased from 118 g/l before to 140 g/l at 12 months and to 146 g/l at 24 FIGURE 2 Changes of MBL before and after LNG-IUS insertion. Changes of MBL of each individual are expressed in the linear graphs, with dots indicating the MBL value at pretreatment cycles ( 2 months and 1 month) and at 6, 12, 24, and 36 months of follow-up. MBL decreased markedly at 6 months in all women. In one case (---), MBL was very high before treatment (259.6 and 150 ml) and decreased to 100 ml at 12 months; thus, it remained higher than the limit of menorrhagia, but the patient became amenorrheic after 24 months. 966 Xiao et al. Treatment of menorrhagia with LNG-IUS Vol. 79, No. 4, April 2003

TABLE 2 Hemoglobin (g/l) before and after levonorgestrel-releasing system insertion. Postinsertion Preinsertion 3 months 6 months 12 months 24 months 36 months No. of patients 34 29 27 29 26 24 Mean 121.5 130.3 132 133.4 138.7 133.5 SD 10.2 9.5 9.8 8.7 7.5 6.6 Maximum 138 149 147 147 150 150 Minimum 96 112 107 111 120 118 P.0001.0001.002.001.0001 Note: The paired t-test was used for comparing hemoglobin data at each segment month with the preinsertion value. The increase of hemoglobin were very significant after 3 months of treatment. months. Serum ferritin increased from 18.3 ng/ml to 29.1 ng/ml at 12 months and to 89.4 ng/ml at 24 months. No fibroids were found in this patient nor were any other causes detected. Subjectively, this patient was very satisfied with the treatment. Serum Levels of E 2, P, and LNG After 3 4 Years of Use Twenty blood samples were collected from patients using the LNG-IUS over 3 4 years. Serum E 2, P, and LNG levels were measured by RIA. The E 2 levels varied between 72 and 1,156 pmol/l, with an average of 239.4 pmol/l. The P levels were mostly low, 11.1 nmol/l on average, but six samples indicated luteal function (13.5 67.0 nmol/l). In eight samples, both P ( 5 nmol/l) and E 2 levels ( 200 pmol/l) were low. Other samples indicated good follicular activity and luteal function. Serum LNG levels were maintained at an average level of 511 pmol/l, but five patients showed exceptionally high levels of over 1,000 pmol/l. DISCUSSION The therapeutic effect of the LNG-IUS in the treatment of menorrhagia is very well illustrated in European studies by Luukkainen et al. (5) and Anderson and Rybo (6). Tang and Lo (11), in 1995, reported their experience in the treatment of menorrhagia in 10 Chinese women for a total of 129 cycles, with significant reductions of MBL of 54%, 87%, and 95% at the first, third, and sixth month of treatment, respectively. In the present study, we report the follow-up of 34 subjects for 3 years, which is the longest follow-up of menstrual pattern after LNG-IUS insertion ever reported in the literature. This implies that the LNG-IUS is very well accepted by Chinese women. More than one-third of the women who had amenorrhea or very scanty bleeding found the LNG-IUS very satisfactory. They had previously been suffering from the bleeding both physically and mentally, with no effective treatment for a long time. They were happy to have amenorrhea, knowing that it was not pathological and was reversible. The mechanism of action has been studied by many authors. The therapeutic effect depends mainly on the direct action of progestin on the endometrium (12 14) and less on the systemic effect on hypothalamo-pituitary-ovarian function (15 17). Hence, the LNG-IUS has advantages over treatment with oral steroid pills or long-acting injectables of TABLE 3 Serum ferritin (ng/ml) before and after levonorgestrel-releasing system insertion. Postinsertion Preinsertion 3 months 6 months 12 months 24 months 36 months No. of patients 33 34 27 28 25 23 Mean 21.9 23.7 33.9 52.8 83.3 92.8 SD 16.7 14.4 17.3 31.5 34.9 50.3 Maximum 76.4 64.7 76.2 133.7 153.2 237.3 Minimum 10.2 9.3 9.1 11.9 13.3 17.9 P.347.0001.0001.0001.0001 Note: The paired t-test was used for comparing serum ferritin data at each segment month with the preinsertion value. The increase of serum ferritin was very significant after 6 months of treatment. FERTILITY & STERILITY 967

progestin. Instead of intermittent fluctuating levels, as with daily pills, or very high systemic levels of steroids, as with steroid injectables, the LNG-IUS provides a local effect with a low systemic level of progestin throughout a long period of several years. This mechanism of action is the basis of the therapeutic effect of the LNG-IUS in the treatment of menorrhagia. The serum levels of E 2 and P after 3 4 years of use in this study also confirm the results of previous studies, showing that ovarian function is little affected after longterm use (16), and good luteal function was maintained in some cases. As the blood samples were taken only once and at random after completion of the study, it is difficult to judge whether or not those with low P levels were in the follicular phase. However, the E 2 levels showed that follicular activity was normal, some even showing periovulatory high levels. Serum LNG concentrations were maintained at almost constant levels, as we have reported in a previous study (16), which indicates that the system could be used for a much longer period than 5 years. Prolonged and irregular spotting and bleeding, although in low amounts, during the early months of treatment is the main disadvantage of the LNG-IUS. Patients with menorrhagia are satisfied with reduced bleeding and are not concerned about the side effects. Prolonged spotting or a small amount of bleeding is more tolerable than heavy bleeding every month. In addition, along with the reduction of MBL, the LNG-IUS has the beneficial effect of increasing blood hemoglobin and whole body iron reserves. Although most patients were used to a low level of iron during the long duration of heavy bleeding, they were not in a healthy condition and additional diseases would worsen their condition. Hence, an increase in body iron reserves is important to women s health in general. For years, these patients had been tired and depressed. It is a great relief to such women when scanty or no bleeding occurs. Regarding younger women, they do not have to worry about accidental pregnancy because they know that the LNG-IUS is also a good contraceptive. Some women experienced some minor side effects, such as lower abdominal pain, headache, backache, and breast tenderness, which were negligible when compared with the previous heavy bleeding. In addition, treatment with the LNG-IUS saved the cost, both physically and financially, of surgical intervention, which had been the usual solution when medical treatment failed. As reported by Lähteenmäki et al. (18), in a study of 56 menorrhagic women waiting for hysterectomy, 64.3% of the women with the LNG-IUS for 6 months cancelled their decision to undergo surgery compared with 14.3% in the control group. There is no doubt that women would be happy to use a device that can stop bleeding as a conservative alternative to hysterectomy. Taking into consideration the total costs of hospitalization, operation, and the postoperative recovery period, one LNG-IUS that can last over 5 years is indeed an excellent method of treatment of menorrhagia. Hurskainen et al. (19) reported the results of a randomized study comparing the LNG-IUS (119 patients) with hysterectomy (117 patients) in the treatment of menorrhagia in terms of the quality of life and cost-effectiveness. They found that the health-related quality of life improved significantly in both the IUS and hysterectomy groups, as did other indices of psychological well-being. Overall, costs were about three times higher for the hysterectomy group than for the IUS group. The result may not be the same in China or in other developing countries. The cost of hysterectomy may be 10 times higher, including all the expenses of hospitalization. For a patient of low socioeconomic status it would be a great burden to the whole family. Even for a minor operation such as endometrial ablation there is no comparison with LNG-IUS insertion as regards cost and safety (20). Hence, a device that can be an alternative to surgical intervention will certainly be welcomed by many women, especially those in developing countries, where medical care and operations are not available or affordable. The cases of total and partial expulsion of the IUS in this study could be due to one of two factors: heavy bleeding and uterine contraction that causes flushing out of the device or descent of the IUS when withdrawing the inserter during the insertion procedure. Therefore, special attention must be paid to proper training of service providers in the insertion technique, and women must be warned of the possibility of expulsion when MBL is over 150 200 ml. Checking at the clinic if bleeding is still heavy after the first or second menstrual period is necessary, and a new IUS could be inserted if expulsion has occurred. Some women may not be willing to tolerate the irregular bleeding that sometimes occurs in the initial phase after LNG-IUS insertion. Hence, counseling before insertion and during follow-up visits about the possibility of irregular bleeding can reassure the patient that bleeding will eventually stop and that reduction of bleeding is beneficial to her health in general. Among the treated women, there were five cases with small fibroids, either subserous or intramural, as detected by ultrasound. MBL in these subjects became markedly reduced, but no decrease in size of the fibroids was detected. The therapeutic effect of the LNG-IUS in treatment of fibromyoma or endometriosis needs to be explored in the future. The contraceptive and therapeutic effects of the LNG-IUS are important to the improvement of reproductive health of women, particularly those in developing countries. Unfortunately, owing to the limited provision of the LNG-IUS in developing countries, the best features have not been well 968 Xiao et al. Treatment of menorrhagia with LNG-IUS Vol. 79, No. 4, April 2003

recognized. More information should be distributed and clinical trials carried out. Acknowledgments: The authors acknowledge the financial support and donation of LNG-IUSs from Leiras Oy, Turku, Finland, and Schering AG, Berlin, Germany. The authors thank Taru Blom, M.D., Ph.D., Leiras Pharmaceutical Co., Turku, Finland, and Pekka Lähteenmäki, M.D., Ph.D., Institute of Biomedicine, University of Helsinki, Finland, for scientific advice; Juhani Tuominen, Ph Lic, University of Helsinki, Finland, for comments and advice; Zhao Heng, BS, Beijing Medical University, for technical assistance in the statistical analysis; and Nicholas Bolton, Ph.D., for revising the language of the manuscript with great expertise. References 1. Luukkainen T, Allonen H, Haukkamaa M, Lähteenmäki P, Nilsson CG, Toivonen J. Five years experience with levonorgestrel releasing IUD. Contraception 1986;33:139 48. 2. Wang SL, Wu SC, Xia XM, Chen JH, Gao J. Three years experience with levonorgestrel-releasing intrauterine device and Norplant-2 implants: a randomized comparative study. Adv Contraception 1992;8: 105 14. 3. Xiao BL, Gu SJ, Wang SL, Zhu PD, Shi SQ. Norplant and the levonorgestrel IUD in Chinese Family Planning Program. Proceedings of the International Symposium on Recent Advances in Female Reproductive Health Care. 5 6 June 1992, Helsinki, Finland. Ann Med 1993;25:161 5. 4. Xiao BL, Zhu PD, Wu SC. Use of levonorgestrel-releasing intrauterine system on contraception and treatment of menorrhagia the Chinese experience. Gynecol For 1998;3(3):26 8. 5. Luukkainen T, Lähteenmäki P, Toivonen J. Levonorgestrel-releasing intrauterine device. Ann Med 1990;22:85 90. 6. Andersson JK, Rybo G. Levonorgestrel-releasing intrauterine device in the treatment of menorrhagia. Brit J Obstet Gynaecol 1990;97:690 4. 7. Hallberg L, Nilsson L. Determination of menstrual blood loss. Scand J Clin Lab Invest 1964;16:244 8. 8. Sufi SB, Donaldson A, Jeffcoat SL. WHO matched reagent program method manual. 19th ed. London, UK: WHO Collaborating Centre for Immunoassay, 1995:25 46. 9. Ahsan R, Sufi SB, Cekan S. Laboratory method manual for the radioimmunoassay of levonorgestrel, norethisterone, and medroxyprogesterone acetate. 3d ed. London, UK: WHO Collaborating Centre for Immunoassay, 1995:19 32. 10. Wilcoxon F. Individual comparisons by ranking methods. Biometrics Bull 1945;1:80 3. 11. Tang GWK, Lo SST. Levonorgestrel intrauterine device in the treatment of menorrhagia in Chinese women: efficacy versus acceptability. Contraception 1995;51:231 5. 12. Nilsson CG, Haukkamaa M, Vierola H, Luukkainen T. Tissue concentration of levonorgestrel in women using a levonorgestrel-releasing IUD. Clin Endocrinol 1982;17:529 36. 13. Zhu PD, Luo HZ, Shi WL, Wang JD, Cheng J, Xu RH, et al. Observation of the action of factor VIII in the endometrium of women preand post-insertion of three types of IUDs. Contraception 1991;44:367 84. 14. Gu Z, Zhe PD, Luo HZ, Zhu XH, Zhang GH, Wu SC. A morphometric study on the endometrial activity of women before and after one year with LNG-IUD in situ. Contraception 1995;52:57 61. 15. Xiao BL, Zhou LY, Zhang XL, Jia MC, Luukkainen T, Allonen H. Pharmacokinetic and pharmacodynamic studies of levonorgestrel-releasing intrauterine device. Contraception 1990;41:353 62. 16. Xiao BL, Zeng T, Wu S, Xiao N. Effect of levonorgestrel releasing intrauterine device on hormonal profiles and menstrual pattern after long term use. Contraception 1995;51:359 65. 17. Barbosa I, Bakos O, Olsson S-E, Odlind V, Johansson EDB. Ovarian function during use of a levonorgestrel-releasing IUD. Contraception 1990;42:51 66. 18. Lähteenmäki P, Haukkamaa M, Puolakka J, Riikonen U, Sainio S, Suvisaari J, et al. Open randomized study of use of lenvonorgestrel releasing intrauterine system as alternative to hysterectomy. Br Med J 1998;316:1122 6. 19. Hurskainen R, Teperi J, Rissanen P, Aalto A-M, Grenman S, Kivela A, et al. Quality of life and cost effectiveness of levonorgestrel-releasing intrauterine system versus hysterectomy for treatment of menorrhagia: a randomised trial. Lancet 2001;357:273 7. 20. Istre O, Trolle B. Treatment of menorrhagia with the levonorgestrel intrauterine system versus endometrial resection. Fertil Steril 2001;76: 304 9. FERTILITY & STERILITY 969