Indications and reasons for discontinuing the levonorgestrel-releasing intrauterine system (LNG-IUS)

Objectives: to identify the main indications and re asons for discontinuing the use of the Levonorgestrel-Releasing Intrauterine System (LNG-I US).

a cross-sectional study was carried out fr om medical records of 327 women who used the LNG-IUS 52mg between January 2011 and December 2016 at a public hospital in the Northeast of Brazil.

the main indications for the use of the LN G-IUS were: contraception (32.7%), myoma/metrorrhagia (28.7%) and endometriosis/pelvic pain (22.3%). Of the 327 women, 68 (20.8%) had discontinued using the device. The most frequent reasons for discontinuation were: expulsion (9.2%), LNG-IUS expiration (3.7%), bleeding (2.4%) and pain (1.5%). Most patients had no difficulty in the insertion and did not require anesthesia/sedation. Among the 30 women who expelled the device, 17 (56.7%) had us ed it for metrorrhagia and myoma, 8 (26.7%) for contraception, and 5 (16.6%) for endome triosis/pelvic pain.

the LNG-IUS is a well-accepted contrac eptive method, with therapeutic applications for some gynecological conditions and a low expulsion rate. Key words Contraception, Contraceptive effectiveness, Levonor gestrel, Long-acting reversible contraception, Progesterone-releasing in trauterine devices This article is published in Open Access under the Creative Commons Attribution license, which allows use, distribution, and reprod uction in any medium, without restrictions, as long as the original work is correctly cited. http://dx.doi.org/10.1590/1806-93042020000200009 Maria Luíza Lemos Pires 1 https://orcid.org/0000-0003-0615-7669 Ariani Impieri Souza 2 https://orcid.org/0000-0002-7917-5983 Maria Luisa Borges Roriz Dantas 3 https://orcid.org/0000-0001-7829-8204 Gabriela Delgado Soriano 4 https://orcid.org/0000-0002-3360-8317 1,3,4 Faculdade Pernambucana de Saúde. Recife, PE, Brasil . 2,5,6 Departamento de Ginecologia e Obstetrícia. Institut o de Medicina Integral Professor Fernando Figueira. Rua dos Coelhos, 300. Boa Vista. Recife, PE, Brasi l. CEP: 50.070-550. E-mail: [email protected] Cláudia Viana Henriques 5 https://orcid.org/0000-0002-7165-543X Ana Laura Carneiro Gomes Ferreira 6 https://orcid.org/0000-0001-9172-6162 Rev. Bras. Saúde Mater. Infant., Recife, 20 (2): 47 9-484 abr-jun., 2020480

The Levonorgestrel-Releasing Intrauterine System (LNG-IUS) was initially developed as a device for contraception. Its use is now recommended by the World Health Organization (WHO), and it is consi- dered one of the most effective long-acting reversible contraception (LARC) methods. 1,2 In the mid-1990s, the LNG-IUS was proven to be an effective treatment for menorrhagia and dysmenorrhea owing to its inhibitory effect on endometrial actions with consequent atrophy and a decrease in the thickness of the uterine endometrium. 3,4 Non-contraceptive benefits of the LNG-IUS, including treatment of abnormal uterine bleeding, were reported in a systematic review conducted in 2015. 5 Currently the LNG-IUS is recommended as the first-line treatment for menorrhagia in the Uni ted Kingdom, and it is approved in 120 countries as a contraceptive method and in 115 countries to contro l menorrhagia. 6 Among other non-contraceptive uses, other researchers have reported proven benefits of the LNG-IUS in the management of dysmenorrhea by decreasing pelvic pain, premenstrual syndrome (PMS) and to treat endometriosis and adeno- myosis.7,8 On the basis of the WHO eligibility criteria, the LNG-IUS can be used safely by both multiparous and nulliparous women, as well as by women in other clinical situations. It is contraindicated in cases of suspected or confirmed breast cancer, malignant disease in the cervix, malformation or distortion o f the uterine cavity, active pelvic infection, suspec ted or confirmed pregnancy, and uterine bleeding of unknown etiology. 8,9 Complications associated with the insertion of the LNG-IUS include uterine perforation, cervix laceration, inadequate device placement, and expul- sion of the device. 10 In addition to complications, other situations that make it difficult or impossib le for LNG-IUS insertion include: stenosis of uterine cervix, distortion of the uterine cavity and pain. 9 The postpartum period while women are still breast- feeding, history of cesarean delivery, and inexperi - ence of the health professionals can also add to th e previously mentioned difficulties. 10 The reasons for the removal of the LNG-IUS include changes related to menstrual bleeding pattern, and although less frequent, hormonal changes causing such as weight gain, acne and mastalgia. 11,12 Expulsion is another reason for discontinuation of the use of the LNG-IUS. 13 In addition to expanding the contraceptive options available to women seeking effective, reversible, and safe methods, the LNG-IUS has been very cost-effective method. 14,15 The study by Pepe et al .15 showed that the LNG-IUS could be cost- effective when compared with other hormonal methods. The present study aims to identify the mai n indications to use the LNG-IUS 52mg and the main reasons for discontinuation of use.

A cross-sectional observational study was carried o ut at the Instituto de Medicina Integral Prof. Fernand o Figueira (IMIP). The medical records of 327 women who used the LNG-IUS between January 2011 and December 2016 were analyzed. We collected the following information from the patients’ medical records: sociodemographic vari- ables (education, residence place, profession, and marital status), biological variables (date of birt h, race, and disease status), reproductive/gynecologic al data (age of menarche, age at first intercourse, number of sexual partners, number of gestations, births, abortions and types of births) and LNG-IUS variables (date of insertion, physician who performed the insertion, indications, any difficult y in insertion, use of anesthesia/sedation, and continued use of the device or reason for discontin u- ation). The data was entered into a database created on Microsoft Excel. The distribution of absolute and relative frequencies was performed by program R, version 3.3.1. Pearson's chi-square test was applie d to verify association between two variables. A p- value <0.05 was considered significant. This project was approved by the Human Research Ethics Committee of IMIP, with CAAE No. 59253516.5.0000.5201.

The mean age of the 327 women at the time of the LNG-IUS insertion was 35 ± 8 years, and 107 (32.7%) had 8 to 11 years of schooling. The majorit y of the women (n=256; 78.3%) came from the Metropolitan Region of Recife and 168 (51.4%) were married or in a stable union. Only 56 (17.1%) were nulliparous and among the parous, 153 (46.8%) had a previous cesarean section (Table 1). Among the main indications to use the LNG-IUS device, 107 women (32.7%) chose this method for contraception; 94 (28.7%) women chose for treating myoma and/or metrorrhagia; 73 (22.3%) for endometriosis associated with pelvic pain, and 27 Souza AI et al. Rev. Bras. Saúde Mater. Infant., Recife, 20 (2): 47 9-484 abr-jun., 2020 481 Indications and reasons for discontinuing the LNG-I US (8.3%) for hematological disease. Other indications (n=26; 8%), included post-abortion contraception, women with renal transplantation, and PMS. Sixty eight (20.8%) women discontinued the use of the LNG-IUS. Thirty (9.2%) of them discontinued by expulsion, 8 (2.4%) discontinued for bleeding an d 5 (1.5%) for pelvic pain. Other reasons for discon- tinuation (n=13; 4.0%) included weight gain, acne and desire to have children. Finally, 12 women (3.7%) wished to change the device at the end of LNG-IUS lifespan. Among the 30 women who expelled the device, 17 (56.7%) had used it for me- trorrhagia and myoma treatment (Table 2). Difficulty in the insertion of the LNG-IUS occurred in 49 (15.0%) women and only 15 (4.6%) women were required anesthesia/sedation. Insertion difficulty was significantly associated with nulli- parity ( p = 0.002) and the need for anesthesia/seda- tion ( p = 0.006). There was no association between the insertion difficulty and type of delivery (p=0.060) (Table 3). Table 1 Distribution of biological and sociodemographic cha racteristics among the women using the LNG-IUS. Variables N=327 % Age between 30-39 years old (mean age - 35 ± 8 yea rs) 137 41.9 Schooling between 8-11 years 107 32.7 Living at Metropolitan Region of Recife 256 78.3 Married / stable union 168 51.4 Nulliparous 56 17.1 Parous 269 82.2 Previous cesarean section 153 46.8 LNG-IUS = Levonorgestrel-Releasing Intrauterine Sys tem. Table 2 Distribution of the main indications to use the LNG -IUS and the main reasons for discontinuation of th e LNG-IUS. Main indications to use N=327 % Contraception 107 32.7 Metrorrhagia / myoma 94 28.7 Endometriosis / pelvic pain 73 22.3 Hematological disease 27 8.3 Other indications 26 8.0 Main reasons for discontinuation Expulsion 30 9.2 Bleeding 8 2.4 Pelvic pain 5 1.5 End of LNG-IUS lifespan 12 3.7 Other reasons 13 4.0 LNG-IUS = Levonorgestrel-Releasing Intrauterine Sys tem. Rev. Bras. Saúde Mater. Infant., Recife, 20 (2): xx x-xxx abr-jun., 2020482 Souza AI et al. Table 3 Association between difficulty in insertion and par ity, type of delivery and necessity of anesthesia/s edation. Variables Difficulty in insertion Total p Yes No n % n % n % Parity 0.002 Nulliparous 16 28.6 40 71.4 56 100.0 Parous 33 10.1 236 72.6 325 100.0 Type of delivery 0.060 Vaginal 7 7.0 91 93.0 98 100.0 Previous cesarean section 23 15.0 130 85.0 153 100.0 Necessity of anesthesia/sedation 0.006 Yes 6 40.0 9 60.0 15 100.0 No 43 14.0 267 86.0 310 100.0

The women in this study had sociodemographic and reproductive characteristics commonly found in studies evaluating long-term contraceptive methods, as the LNG-IUS. 10,16,17 At the time of LNG-IUS insertion, their mean age was 35 (± 8) years old and had between 8-11 years of schooling. A study that followed 61,448 patients over seven years in six European countries recorded a mean age of 37.4 years among women who opted for the LNG-IUS. 10 Probably because women in reproductive age group are searching for contraceptive methods with high efficacy and safety.14,18 In addition, women with higher levels of schooling who are more informed about the safety and efficacy of LARCs and may be more motivated to choose this type of contraception in order to av oid an unplanned pregnancy. Most of the women in this study had eight or more years of schooling, and thi s level of schooling is similar to PNAD ( Pesquisa Nacional por Amostra de Domicílio /National Household Sample Survey) data, which showed an improvement in the educational level in the Brazili an population. 19 According to the CHOICE contracep- tive project, which included 9,256 women from St Louis region, in the United States (USA), women with less education and financial restrictions had greater difficulty in obtaining contraceptive methods.18 Although contraception was the most frequent indication for the insertion of the LNG-IUS, other indications are described as high efficacy of the device in the reduction of bleeding and gynecolo- gical pathologies treatment. 5 In the present study, some women used the method to treat uterine bleeding, myoma and adenomyosis. Endometriosis associated with pelvic pain and hematological disease were also the reasons to insert the device, according to other studies. 5,7,17 The use of the LNG-IUS for the treatment of endometriosis associated with pelvic pain is well documented. This treatment is effective in relievin g pain, leading to an improvement in the quality of l ife of women with endometriosis. 5,7,8 The expansion of the indications for the LNG-IUS insertion is due to its mechanism of action, which promotes endome- trial atrophy with consequent control of uterine bleeding and pain. 3 On the other hand, the LNG-IUS device may present adverse effects that lead women to discontinue the method. 20 Among the 327 women in this study, 68 (20.8%) discontinued the use of the device. The main reason for discontinuation was the device expelled sponta- neously (9.2%), similar to a study performed with 481 women in the USA whose percentage of sponta- neous expulsion was 9.6% 16 and another study carried out in the USA involving 5,403 women using the device with an expulsion rate of 10.1%. 21 A 7- year randomized controlled trial in 20 centers in n ine countries found a lower expulsion rate (8.2%). 22 It is interesting to observe that this large WHO study 22 was a contraceptive trial where all the women involved had used the LNG-IUS for contraception while in our study there were patients using LNG- IUS for gynecological pathologies treatment which may favor the expulsion. Of the thirty women who expelled the LNG-IUS in our study, the majority had inserted the device for treating metrorrhagia/myoma. This is in line with a study that found a significantly high rate of spont a- neous expulsion of the device in women with uterine pathologies. 21 Rev. Bras. Saúde Mater. Infant., Recife, 20 (2): 47 9-484 abr-jun., 2020 483 Indications and reasons for discontinuing the LNG-I US Discontinuation for bleeding-related complaints occurred in 8 (2.4%) LNG-IUS users. This rate was higher than a trial that assessed 3-year data on th e efficacy and safety of a LNG-IUS conducted at 29 clinical sites in the USA (1.5%). 23 Besides this, the large WHO study 22 found a cumulative rate of method discontinuation due to bleeding problems higher in the LNG-IUD than the TCu380A group. And also, the increased bleeding volume and frequency were associated with reduced short-term satisfaction in LNG-IUS users among participants. 24 Most of the women in our study were multi- parous, coinciding with data from Youm et al .16 study, in which multiparous women were the majority in the sample. This may be explained because in the past intrauterine devices have been recommended only for women who have already had children. The LNG-IUS expanded the indications for intrauterine devices, including contraception for nulliparous and this indication became more frequently and supported by scientific evidence.9,11,25 Parity has also been considered a factor that may be related to the difficulty in insertion of the LN G- IUS9,10 although some studies reported no difficulty in the insertion in nulliparous. 1,11,25 In the present study, the nulliparous tended to experience greater difficulty with the insertion of the LNG-IUS. One explanation for this could be because for many years, the insertion of a copper IUD in nulliparous was contraindicated, 11 which has led to a lower ability of the professionals for inserting the LNG- IUS in this group of women. In this study there was no association between difficulty with insertion of the device and women with previous cesarean section. It is known that th e area of scar tissue changes in the isthmus region after cesarean section may hinder the passage of th e device during insertions. Although a history of cesarean section is not contraindicated for the LNG - IUS insertion, 11 a study compared the difficulty of insertion among women with and without a previous cesarean section found that most of the women who presented difficulty in insertion had a history of cesarean delivery. 9 It is important to consider the limitations of the present study. Since it is a cross-sectional retros pec- tive study, some medical records were incomplete or illegible. This study concluded that the main indication for the use of the LNG-IUS was contraception. Most of the women had good adherence to the method, had no difficulty in insertion and did not require anes - thesia / sedation. Randomized and controlled prospective studies evaluating a greater number of women using the LNG-IUS, especially in public hospitals and clinics may produce robust scientific evidence. These

could alert public health managers about th e benefits of LNG-IUS in the public service and improve women’s health care. Author’s contribution Pires MLL, Souza AI, Henriques CV and Ferreira ALCG contributed to the conception and design, analysis, interpretation of the data, and writing o f the article. Pires MLL, Dantas MLBR, Soriano GD and Henriques CV contributed to the data collection, analysis, and writing of the article. Souza AI and Ferreira ALCG also contributed to the critical revi ew of the intellectual content. All the authors have approved the final version to be published.

1. Sabbioni L, Petraglia F, Luisi S. Non-contracepti ve benefits of intrauterine levonorgestrel administration: why not? Gynecol Endocrinol. 2017; 33: 822-9. 2. WHO (World Health Organization). Family Planning: A global handbook for providers. 2018 Edition. [Cited 10 Jul 2018]. 2018. Available from: http://www.who.int/rep roduc- tivehealth/publications/fp-global-handbook/en/ 3. Luo L, Luo B, Zheng Y, Zhang H, Li J, Sidell N. Levonorgestrel-releasing intrauterine system for at ypical endometrial hyperplasia. Cochrane Database Syst Rev . 2013; 6: CD009458. 4. Cruz J, Costa AR. Levonorgestrel-releasing intrau terine system and breast cancer. Acta Obstet Ginecol Port. 2017; 11 (4): 276-82. 5. Bahamondes L, Bahamondes MV, Shulman LP. Non- contraceptive benefits of hormonal and intrauterine reversible contraceptive methods. Hum Reprod Update . 2015; 21: 640-51. 484 Rev. Bras. Saúde Mater. Infant., Recife, 20 (2): 47 9-484 abr-jun., 2020 Souza AI et al. 6. Qiu J, Cheng J, Wang Q, Hua J. Levonorgestrel-rel easing intrauterine system versus medical therapy for meno r- rhagia: a systematic review and meta-analysis. Med Sci Monit Basic Res. 2014; 20: 1700-13. 7. Park DS, Kim M La, Song T,. Yun BS, Kim MK, Jun H S, Seong SJ. Clinical experiences of the levonorgestre l- releasing intrauterine system in patients with larg e sympto- matic adenomyosis. Taiwan J Obstet Gynecol. 2015; 5 : 412-5. 8. WHO (World Health Organization). Medical eligibil ity criteria for contraceptive use. 5th ed. 2015. WHO L ibrary Cataloguing-in-Publication Data [Cited 21 jul 2017] . Available from: https://apps.who.int/iris/bitstream/handle/10665/18 1468/97 89241549158_eng.pdf 9. Santos AR, Bahamondes MV, Hidalgo MM, Atti A, Bahamondes L, Monteiro I. Pain at insertion of the levonorgestrel-releasing intrauterine system in nul ligravida and parous women with and without cesarean section. Contraception. 2013; 88: 164-8. 10. Heinemann K, Reed S, Moehner S, Do Minh T. Risk of uterine perforation with levonorgestrel-releasing a nd copper intrauterine devices in the European Active Surveillance Study on Intrauterine Devices. Contrac eption. 2015; 91: 274-9. 11. Carvalho NS, Botelho AB, Mauro DP, Ferreira KA, Amaro LC, Mendes PC, Savian RR, Carvalho BF. Sexually Transmitted Infections, Pelvic Inflammatory Disease , and the role from Intrauterine Devices: myth or fact? J Biomedical Sci. 2016; 6: 1. 12. Nascimento R, Hobus D, Jesuíno F, Figueiredo J, Bahamondes L. Evaluation of quality of life and pai n control in patients with endometriosis-related pain after insertion of the levonorgestrel-releasing intrauter ine system (LNg-IUS). Arq Catarin Med. 2013; 42: 21-7. 13. Liang H, Li L, Yuan W, Zou Y, Gao ES, Duffy JM, Wu SC. Dimensions of the endometrial cavity and intrauteri ne device expulsion or removal for displacement: a nes ted case–control study. BJOG. 2014; 121: 997-1004. 14. Secura GM, McNicholas C. Long-acting reversible contra- ceptive use among teens prevents unintended pregnan cy: a look at the evidence. Expert Rev Obstet Gynecol. 20 13; 8: 297-9. 15. Pepe C, Sakagute M, Ramos N, Maestri CA. Cost-ef fective- ness analysis of 52 mg levonorgestrel-releasing int rauterine system (LNG-IUS) versus short-acting hormonal contr a- ceptives in avoiding unintended pregnancy (UP) in 1 5 to 19 year-old adolescents under the Brazilian Public Hea lthcare System (SUS). J Bras Econ Saúde. 2017; 9 (1): 100-8 . 16. Youm J, Lee HJ, Kim SK, Kim H, Jee BC. Factors a ffecting the spontaneous expulsion of the Levonorgestrel-rel easing intrauterine system. Int J Gynaecol Obstet. 2014; 1 26: 165- 9. 17. Bahamondes MV, Hidalgo M, Bahamondes L, Monteiro I. Ease of insertion and clinical performance of the levonorgestrel-releasing intrauterine system in nulligravidas. Contraception. 2011; 84: e11-16. 18. Secura GM, Allsworth JE, Madden T, Mullersman JL , Peipert JF. The Contraceptive CHOICE Project: Reduc ing Barriers to Long-Acting Reversible Contraception. A m J Obstet Gynecol. 2010; 203: 115.e1–115.e7. 19. Brazil. IBGE (Insituto Brasileiro de Geografia e Estatística). Pesquisa nacional por amostra de domi cílios: síntese de indicadores; 2015 / Coordenação de Traba lho e Rendimento. [National survey by sample of household s: summary of indicators. 2015 / IBGE, Coordination of Work and Income] Rio de Janeiro: IBGE; 2016. [Cited 25 J ul 2018]. Available from URL:https://biblioteca.ibge.gov.br/visualizacao/liv ros/liv98 887.pdf 20. Van der Heijden PAHH, Geomini PMAJ, Herman MC, Veersema S, Bongers MY. Timing of insertion of levonorgestrel-releasing intrauterine system: a ran domised controlled trial. BJOG. 2017; 124: 299-305. 21. Madden T, McNicholas C, Zhao Q, Secura GM, Eisen berg DL, Peipert JF. Association of age and parity with intrauterine device expulsion. Obstet Gynecol. 2014 ; 124: 718-26. 22. Rowe P Farleya T, Peregoudova A, Piaggiob G, Boccarda S, Landoulsic S, Meirikd O, and the IUD Research Group of the UNDP/UNFPA/WHO/World Bank Special Programme of Research. Safety and efficacy in parous women of a 52- mg levonorgestrel-medicated intrauterine device: a 7-year randomized comparative study with the TCu380A. Contraception. 2016; 93: 498-506. 23. Eisenberg DL, Schreiber CA, Turok DK, Teal SB, W esthoff CL, Creinin MD. Three-year efficacy and safety of a new 52-mg levonorgestrel-releasing intrauterine system. Contraception. 2015; 92: 10-6. 24. Dietrish JT, Desai S, Zhao Q, Secura G, Madden T , Peipert JF. Association of Short-term Bleeding and Cramping Patterns with Long-Acting Reversible Contraceptive

Satisfaction. Am J Obstet Gynecol. 2015; 212 (1): 50.e1–50.e8. 25. Lohr PA, Lyus R, Prager S. Use of intrauterine d evices in nulliparous women. Contraception, 2017; 95: 529-53. ______________ Received on November 20, 2019 Final version presented on February 20, 2020 Approved on March 30, 2020