D 1205 OULU 2013 D 1205

UNIVERSITY OF OULU P.O.B. 7500 FI-90014 UNIVERSITY OF OULU FINLAND ACTA UNIVERSITATIS OULUENSIS ACTA UNIVERSITATIS OULUENSIS ACTA

SERIES EDITORS DMEDICA Sari Ahonkallio

ASCIENTIAE RERUM NATURALIUM Sari Ahonkallio Senior Assistant Jorma Arhippainen ENDOMETRIAL THERMAL BHUMANIORA ABLATION University Lecturer Santeri Palviainen CTECHNICA A CHOICE FOR TREATMENT OF HEAVY Docent Hannu Heusala MENSTRUAL DMEDICA Professor Olli Vuolteenaho ESCIENTIAE RERUM SOCIALIUM University Lecturer Hannu Heikkinen FSCRIPTA ACADEMICA Director Sinikka Eskelinen GOECONOMICA Professor Jari Juga

EDITOR IN CHIEF Professor Olli Vuolteenaho PUBLICATIONS EDITOR

Publications Editor Kirsti Nurkkala UNIVERSITY OF OULU GRADUATE SCHOOL; UNIVERSITY OF OULU, FACULTY OF MEDICINE, ISBN 978-952-62-0142-9 (Paperback) INSTITUTE OF CLINICAL MEDICINE, ISBN 978-952-62-0143-6 (PDF) DEPARTMENT OF OBSTETRICS AND GYNECOLOGY; ISSN 0355-3221 (Print) OULU UNIVERSITY HOSPITAL ISSN 1796-2234 (Online)

ACTA UNIVERSITATIS OULUENSIS D Medica 1205

SARI AHONKALLIO

ENDOMETRIAL THERMAL ABLATION A choice for treatment of

Academic dissertation to be presented with the assent of the Doctoral Training Committee of Health and Biosciences of the University of Oulu for public defence in Auditorium 4 of Oulu University Hospital, on 7 June 2013, at 12 noon

UNIVERSITY OF OULU, OULU 2013 Copyright © 2013 Acta Univ. Oul. D 1205, 2013

Supervised by Docent Markku Santala Professor Hannu Martikainen

Reviewed by Doctor Jyrki Jalkanen Docent Maritta Hippeläinen

ISBN 978-952-62-0142-9 (Paperback) ISBN 978-952-62-0143-6 (PDF)

ISSN 0355-3221 (Printed) ISSN 1796-2234 (Online)

Cover Design Raimo Ahonen

JUVENES PRINT TAMPERE 2013 Ahonkallio, Sari, Endometrial thermal ablation. A choice for treatment of heavy menstrual bleeding University of Oulu Graduate School; University of Oulu, Faculty of Medicine, Institute of Clinical Medicine, Department of Obstetrics and Gynecology, P.O. Box 5000, FI-90014 University of Oulu, Finland; Oulu University Hospital, Kajaanintie 50, FI-90220 OYS, Finland Acta Univ. Oul. D 1205, 2013 Oulu, Finland

Abstract Heavy menstrual bleeding causes significant health and social problems for up to 30% of women at some point of their lives. Medical treatment is not always sufficient or tolerated by women. Hysterectomy is a definitive solution, but it is a major operation associated with long disability and potential severe complications. Endometrial ablation techniques have been developed to avoid the risks related to hysterectomy. Further evolution of these techniques also offers a possibility of a quick and simple outpatient procedure. This study evaluated the long-term effects of endometrial ablation on heavy menstrual bleeding and later endometrial diagnostics. Another aim was to compare the costs when the procedure was performed in different settings. Finally, the effect of hyaluronic acid gel on intrauterine formation was assessed. Endometrial ablation had a good long-term effect on heavy menstrual bleeding in a retrospective study of 172 women, and up to 84% avoided hysterectomy during the follow-up time mean of 5 years. Seventy-six per cent of the patients were satisfied with the procedure. Due to the formation of intrauterine adhesions, prior endometrial ablation compromised later diagnostics of , and outpatient endometrial sampling failed in 23% of 57 women who had undergone endometrial ablation a mean of 6 years earlier, but that did not seem to have clinical importance. In a prospective, randomized and double-blind pilot study of 36 patients, hyaluronic acid gel did not prevent the formation of intrauterine adhesions. In a cost-minimisation analysis based on real resource use, performing endometrial ablation as an outpatient procedure under local anaesthetic instead of a day case procedure performed in the operating theatre under general anaesthetic reduced the costs significantly, from 1,865 to 1,065 euros. In conclusion, the results of this study suggest that endometrial ablation is a good alternative for the treatment of heavy menstrual bleeding, and remarkable cost savings can be achieved by taking the procedure out of the operating theatre. The formation of intrauterine adhesions is common and cannot be prevented with hyaluronic acid gel.

Keywords: cost savings, endometrial ablation techniques, menorrhagia, menstruation disturbances, therapy

Ahonkallio, Sari, Kohdun limakalvon lämpöhoito. Vaihtoehto runsaiden kuukautisten hoitoon Oulun yliopiston tutkijakoulu; Oulun yliopisto, Lääketieteellinen tiedekunta, Kliinisen lääketieteen laitos, Synnytys ja naistentaudit, PL 5000, 90014 Oulun yliopisto; Oulun yliopistollinen sairaala, Kajaanintie 50, 90220 OYS Acta Univ. Oul. D 1205, 2013 Oulu

Tiivistelmä Runsaat kuukautiset aiheuttavat merkittävää terveydellistä ja sosiaalista haittaa jopa kolmasosal- le naisista jossain elämänvaiheessa. Lääkehoito ei aina ole riittävä, eivätkä kaikki naiset voi tai halua käyttää sitä. Kohdunpoisto on lopullinen ratkaisu, mutta se on iso leikkaus, johon liittyy pitkä työkyvyttömyys ja vakavien komplikaatioiden riski. Näiden riskien välttämiseksi on kehi- tetty kohdun limakalvon tuhoavia tekniikoita, joista nykyisin eniten käytetty on limakalvon tuhoaminen lämpöhoidon avulla. Nykytekniikoilla toimenpide voidaan myös tehdä helposti ja nopeasti polikliinisesti. Tässä tutkimuksessa arvioitiin kohdun limakalvon lämpöhoidon pitkäaikaisvaikutuksia run- saiden kuukautisten hoidossa ja sen vaikutusta myöhemmin tapahtuvaan kohdun limakalvon diagnostiikkaan. Niin ikään verrattiin päiväkirurgisen ja polikliinisen toimenpiteen kustannuksia. Lopuksi tutkittiin pystytäänkö hyaluronihappogeelin avulla estämään kohdunsisäisten kiinnik- keiden muodostumista. Lämpöhoidolla oli hyvä pitkäaikaisvaikutus runsaisiin kuukautisiin 172 naista käsittäneessä retrospektiivisessä tutkimuksessa, ja kohdunpoistolta välttyi keskimäärin 5 vuoden seuranta- aikana 84 % naisista. 76 % naisista oli tyytyväisiä hoitoon. Lämpöhoidon aiheuttamat kohdunsisäiset kiinnikkeet vaikeuttivat myöhempää kohdun lima- kalvon diagnostiikkaa. Polikliininen imunäytteen otto ei onnistunut 23 %:lla 57 potilaasta, joille oli tehty lämpöhoito keskimäärin 6 vuotta aikaisemmin. Tällä ei kuitenkaan näyttänyt olevan juurikaan kliinistä merkitystä. 36 potilasta käsittäneessä, prospektiivisessa, satunnaistetussa kak- soissokkotutkimuksessa hyaluronihappogeelin avulla ei pystytty estämään kohdunsisäisten kiin- nikkeiden muodostumista. Todelliseen resurssien käyttöön perustuvassa kustannusten minimointianalyysissa todettiin, että tekemällä lämpöhoito polikliinisesti paikallispuudutuksessa leikkaussalissa nukutuksessa tehtävän toimenpiteen sijasta, kustannukset laskevat 1865 eurosta 1065 euroon. Tämän tutkimuksen perusteella kohdun limakalvon lämpöhoito tarjoaa hyvän vaihtoehdon runsaiden kuukautisten hoitoon, ja sen kustannuksia voidaan merkittävästi pienentää tekemällä toimenpide polikliinisesti. Kohdunsisäisten kiinnikkeiden muodostuminen on tavallista, eikä sitä pystytä estämään hyaluronihappogeelin avulla.

Asiasanat: kohdun limakalvon lämpöhoito, kohdun limakalvon poistomenetelmät, kustannussäästöt, kuukautishäiriöt, menorrhagia, runsaat kuukautiset

To my beloved ones

8 Acknowledgements

This work was carried out at the department of Obstetrics and Gynaecology of the University of Oulu during the years 2005–2013. I wish to express my sincere thanks to Professor Juha Tapanainen, M.D., who was Head of the Department of Obstetrics and Gynaecology for most of that time, for his intelligent and constructive support to my work. Whenever I consulted him, I never left his office empty-handed. I have been privileged to work under the supervision of two tutors whose knowledge and personal properties complement each other. This whole work was initially the idea of docent Markku Santala, who has always been my operative idol, and to whom I owe a debt of gratitude for pretty much all of my skills in that field. Professor Hannu Martikainen was the one who initially introduced me to the world of science, although it took several years before I was mature enough to conduct my own study. I have considered myself extremely lucky to have supervisors with whom there has not been one meeting without humour during these years. I owe a special debt of gratitude to Annikki Liakka, M.D., who has always been so friendly and helpful. With her help I finally understood something about histology. My gratitude goes to Professor Hannu Valtonen from the University of Kuopio for his crucial help in economics. Without his help I would have been totally lost in this area. I want to thank Pirkko Pesonen and Riitta Karjalainen, specialist nurses from the Outpatient Clinic of Gynaecology, who took care of the practical arrangements with such great precision. They never lost their nerves and always looked after practical arrangement. I also want to thank the whole staff in the policlinics, operating theatre and gynaecological ward for an enjoyable co- operation on a daily basis. My special thanks go to my delightful colleagues, who are not only co- workers, but also friends with whom I have the privilege to spend time every day and who have supported me through the tougher days. Laughing with them is the spice of my everyday life. Especially I want to thank Liisa Laatio, M.D., Anna Terho, M.D., Kati Ojala, M.D., Päivi Malmström, M.D., and Sanna Koivunen, M.D., for all the deep conversations concerning life and womanhood.

9 I want to give my warmest thanks to my friend Kirsi Kelhä, M.D., for a friendship that has lasted over half of our lives. We have shared so many things with each other during these years of growing up. My lifetime thanks go to two persons who have known me the longest: my mother Maj-Lis Ahonkallio, who has always helped me and my family altruistically, and my sweet little sister Salla Ahonkallio, with whom I have been fighting and laughing for all her life and almost all my life. The foundation of my life is unquestionably my family. My lovely and lively children Otso, Aarni and Hilla are the most precious gift of my life, and I simply adore them. My heartfelt thanks go to Antti for his loving companionship that has been such a solid rock in my life during all these years. I am extremely happy and grateful for the love and understanding that I have been given, and I hope I will be able to cherish that for the rest of my life. This work was financially supported by Oulu University Research Foundation, the National Graduate School of Clinical Investigation, Astellas Pharma and the Finnish Society of Obstetrics and Gynaecology.

Oulu, May 2013 Sari Ahonkallio

10 Abbreviations

HMB Heavy menstrual bleeding HRT Hormone replacement therapy HTA Hydrothermablation EA Endometrial ablation ELITT Endometrial laser intrauterine thermotherapy LNG-IUS Levonorgestrel-releasing intrauterine system MEA Microwave endometrial ablation PATSS Postablation tubal sterilization syndrome PBAC Pictorial blood loss assessment chart QALY Quality-adjusted life-year RCT Randomized controlled trial REA Rollerball electrosurgical ablation RFEA Radiofrequency endometrial ablation SGY Suomen Gynekologiyhdistys (Finnish Society of Obstetrics and Gynaecology) TBA Thermal balloon ablation TCRE Transcervical resection of the endometrium VAS Visual analogue scale

11

12 List of original publications

This thesis is based on the following articles, which are referred to in the text by their Roman numerals. Some unpublished data are also presented.

I Ahonkallio S, Martikainen H & Santala M (2008) Endometrial thermal balloon has a beneficial effect on menorrhagia. Acta Obstet Gynecol Scand 87: 107–110 II Ahonkallio S, Liakka A, Martikainen H, Santala M (2009) Feasibility of endometrial assessment after thermal ablation. Eur J Obstet Gynecol Reprod Biol 147: 69–71 III Ahonkallio S, Santala M, Valtonen H, Martikainen H (2012) Cost-minimisation analysis of endometrial thermal ablation in a daycase or outpatient setting under different anaesthesia regimens. Eur J Obstet Gynecol Reprod Biol 162: 102–104 IV Ahonkallio S, Martikainen H, Santala M (2012) Auto-crosslinked hyaluronic acid gel in prevention of intrauterine adhesions after bipolar radiofrequency endometrial ablation. A randomised, controlled and double blind pilot study. A manuscript.

13

14 Contents

Abstract Tiivistelmä Acknowledgements 9 Abbreviations 11 List of original publications 13 Contents 15 1 Introduction 17 2 Review of the literature 19 2.1 Prevalence and managing of heavy menstrual bleeding ...... 19 2.2 History and trends of endometrial ablation ...... 20 2.3 Techniques ...... 22 2.3.1 First generation procedures ...... 22 2.3.2 Second-generation procedures ...... 23 2.3.3 Anaesthesia and analgesia ...... 27 2.4 Macroscopic and histopathologic changes after endometrial ablation ...... 29 2.5 Indications and contraindications ...... 30 2.5.1 Endometrial ablation in the presence of myomas or ...... 31 2.6 Complications ...... 32 2.6.1 Intraoperative complications ...... 32 2.6.2 Postoperative complications ...... 33 2.7 Endometrial ablation and fertility ...... 35 2.7.1 Endometrial ablation after previous caesarean delivery ...... 36 2.7.2 Concomitant hysteroscopic sterilization ...... 36 2.8 Outcome ...... 38 2.8.1 Reduction of bleeding ...... 38 2.8.2 Probability of and reasons for hysterectomy ...... 38 2.8.3 Prognostic factors ...... 40 2.8.4 Patient satisfaction ...... 41 2.8.5 Comparison of different second generation techniques ...... 41 2.9 Outcome of endometrial ablation compared to levonorgestrel device ...... 43 2.10 Outcome of endometrial ablation compared to hysterectomy ...... 44 2.11 Endometrial ablation and endometrial cancer ...... 45

15 2.12 Hormone replacement therapy after endometrial ablation ...... 46 2.13 Economics ...... 46 3 Aims of the study 49 4 Materials and methods 51 4.1 Study subjects ...... 51 4.2 Methods ...... 52 4.2.1 Study design ...... 52 4.2.2 Data collection ...... 53 4.2.3 Surgical procedures (Study II and IV) ...... 54 4.2.4 Statistics ...... 54 5 Results 55 5.1 Long-term effects and patient satisfaction ...... 55 5.1.1 Avoiding of hysterectomy...... 55 5.1.2 Bleeding profile and ...... 55 5.1.3 Patient satisfaction ...... 56 5.1.4 Assessment of the endometrium ...... 56 5.2 Cost-minimization ...... 57 5.3 Effect of hyaluronic acid gel in prevention of adhesions ...... 57 6 Discussion 59 6.1 Long-term effects of endometrial ablation ...... 59 6.2 Obliteration of the uterine cavity ...... 60 6.3 Economics ...... 62 7 Conclusions 63 References 65 Original publications 79

16 1 Introduction

Heavy menstrual bleeding (HMB) is defined as excessive menstrual blood loss which interferes with a woman’s physical, social, emotional and/or material quality of life. It is estimated that up to 30% of women suffer from HMB at some point of their lives. Some 20 years ago hysterectomy was often the only treatment offered. Hysterectomy is a major operation which may have some serious adverse effects. The number of hysterectomies has decreased after introduction of less invasive surgical interventions like endometrial ablation and levonorgestrel-releasing intrauterine device (LNG-IUS, Mirena®). Today, LNG-IUS is the first-line treatment option for HMB. However, almost half of the patients treated with LNG-IUS end up with hysterectomy. In many countries endometrial ablation (EA) is the next option after failed LNG-IUS treatment or even an alternative for LNG- IUS. In those countries the use of EA has been increasing over the years, but in Finland it has not reached the same popularity. The guidelines for treatment of menorrhagia were released in Finland in 2005 (The Current Care editorial office 2005, updated 2009) and it recommended the use of EA only if there are contraindications for hysterectomy. Global endometrial ablation (EA) means destroying the endometrium without direct visual control with thermal balloon endometrial ablation (TBA), bipolar radiofrequency endometrial ablation (RFEA), hydrothermablation (HA), microwave endometrial ablation (MEA), or endometrial cryotherapy. These techniques produce less morbidity and a quicker return to work than hysterectomy. The latest devices have very short treatment times, allowing the procedure to be performed as an outpatient procedure under local anaesthetic. EA is suitable for women who have no desire for future pregnancies, and whose is not greatly enlarged or has histological or major structural abnormality. It has been evaluated that EA could effectively replace 30% of hysterectomies performed for HMB. Since research has shown that hysterectomy is a highly effective treatment, these alternatives must be assessed against the recognized high satisfaction rates and improved quality of life reported following hysterectomy. Additional issues that would also need to be addressed include complication rates, side effects and cost- effectiveness. For women with prolonged abnormal uterine bleeding, recent research suggests that hysterectomy is significantly superior to ban expanded medical treatment regimen for health-related quality-of-life measures. The long-term effects of EA have not been studied in Finland before. In this study we evaluated the long-term effects of endometrial ablation on avoiding

17 hysterectomy, patient satisfaction and the effects on later endometrial assessment. We also evaluated the difference in costs when the procedure is taken out of the operating theatre. Finally, we evaluated the effect of hyaluronic acid gel in preventing formation of intrauterine adhesions after EA.

18 2 Review of the literature

2.1 Prevalence and managing of heavy menstrual bleeding

Heavy menstrual bleeding (HMB), previously called menorrhagia, is defined as menstrual periods lasting more than 7 days and/or involving blood loss exceeding 80 ml during one menstrual cycle (Marret et al. 2010). In practice it is however impossible to measure the amount of blood loss precisely, so the diagnosis is often based on a woman’s subjective opinion. Pictorial blood loss assessment charts (PBAC) may also be used. HMB has a major impact on women’s quality of life, and any intervention should rather aim to improve this than focusing on menstrual blood loss (National Institute for Health and Clinical Excellence 2007). Prevalence of HMB is about 22% to 35%, increasing with age, and three women out of ten suffer from HMB at some point of their lives (Shapley et al. 1995, Luoto et al. 2003). In idiopathic HMB (no specific reason for heavy periods and the uterus is normal) the first-line treatment is medical: levonorgestrel-releasing intrauterine device (Mirena®), tranexamic acid, NSAIDs or oral contraceptives. In the early 1990s, it was estimated that at least 60% of women presenting with HMB went on to have a hysterectomy (National Institute for Health and Clinical Excellence 2007), and more than one third of the women undergoing hysterectomy for HMB at that time had an anatomically normal uterus removed (Clarke et al. 1995). Since the 1990s the number of hysterectomies has been decreasing rapidly, very much due to development of levonorgestrel-releasing intrauterine system (LNG- IUS, Mirena®), which was launched 1990. LNG-IUS is effective in reducing menstrual blood loss and is today the primary option for treating HMB in Finland. If a woman is not satisfied with the effect of LNG-IUS, has unwanted adverse effects or does not want to have LNG-IUS at all, she is usually offered hysterectomy. However, hysterectomy is a major operation which causes discomfort and considerable disability in the weeks following surgery, has a 3% rate of severe complications (McPherson et al. 2004, Brummer et al. 2011) and causes minor morbidity (mainly fever and infection) in up to 30% of women. Hysterectomy is thought to be associated with urinary incontinence many years after the operation (van der Vaart et al. 2002), may cause early ovarian failure (Moorman et al. 2011) and has significant cost implications. That is why a conservative surgical technique should be proposed before hysterectomy. The

19 evolution of the surgical treatment of HMB has resulted in the design and manufacture of instrumentation that is minimally invasive, has a low profile risk and is technically simple to operate (Roy & Mattox 2002).

2.2 History and trends of endometrial ablation

Post-traumatic uterine adhesion formation and were first recognized by Fritsch in 1894 (Fritsch 1894). Later Asherman described it as a syndrome (Asherman 1950). Since then, chemical destruction of the endometrium has been attempted using quinacrine, methyl-cyanoacrylate, and paraformaldehyde (Banu & Manyonda 2005). In recent years various surgical techniques have evolved to ablate the endometrium. As a result, the number of hysterectomies in the United Kingdom declined by 64% between 1995 and 2002 (Middleton et al. 2010). In England endometrial ablation is now more common than hysterectomy for HMB (Reid 2007). The National Institute for Health and Clinical Excellence (NICE) recommends in its clinical guideline for heavy menstrual bleeding that if medical treatment fails, endometrial ablation is preferable to hysterectomy if the uterus is not bigger than at 10-week pregnancy and there are no fibroids bigger than 3 cm in diameter (National Institute for Health and Clinical Excellence 2007). Data from UK hospital episode statistics show a significant increase in the overall number of inpatient endometrial ablative techniques (Figure 1). In Finland the number of endometrial ablations performed peaked to 397 in 2005, decreasing after that, being 266 in 2011, which was less than half of the number of hysterectomies (n=695) performed for heavy menstrual bleeding (Figure 2). The slight decline in numbers after 2005 may be due to the Finnish Society of Gynaecology (SGY) guideline for heavy menstrual bleeding, which was published in 2005 and suggested endometrial ablation only if the operative risks of hysterectomy are remarkable (Finnish Society of Gynaecology).

20 Fig. 1. Trends in types of inpatient endometrial ablation procedures performed in England, 2004–11 (data from hospital episode statistics).

1400

1200

1000

800 600

Operations 400

200

0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Hysterectomy Ablation

Fig. 2. Trends in numbers of hysterectomies and endometrial ablations for heavy menstrual bleeding (HMB) 1997–2011 in Finland (data from THL/hoitoilmoitus- järjestelmä).

21 2.3 Techniques

Endometrial ablation techniques can be divided into two broad groups: the first- generation hysteroscopic endometrial ablation procedures performed under direct vision, and the second-generation non-hysteroscopic endometrial ablation procedures which are largely “blind”. The purpose is to remove the entire thickness of the endometrium (inner lining of the uterus). The endometrium has a great tendency of regeneration, and in order to suppress menstruation successfully it is essential to remove the full thickness of this lining together with the superficial (wall of the uterus), including the deep basal glands. The latter are believed to be the primary foci for endometrial regrowth.

2.3.1 First generation procedures

Operative hysteroscopy with transcervical resection has been used since the early 1980s. Goldrath et al. were the first to describe photo vaporization of the endometrium with the neodymium-YAG laser (Goldrath et al. 1981). Soon thereafter, the urologic resectoscope was modified and used for hysteroscopic electrosurgical resection of the endometrium (transcervical resection of the endometrium, TCRE) (DeCherney & Polan 1983, DeCherney et al. 1987, Magos et al. 1989). Rollerball electrosurgical ablation (REA) was first reported by Vancaillie in 1989 (Vancaillie 1989). Traditional resection (or ablation) is performed under regional or general anaesthesia using hysteroscopic guidance and distension media to visualize the endometrial cavity. To avoid loss of current density, a non-ionic, low-viscosity fluid is commonly used. If the pressure created by these fluids overcomes venous pressure in the uterine vascular channels, it may lead to hyponatremic, hypo- osmolar hypervolemia. With prolonged procedures, severe hypotonic fluid overload can result in pulmonary oedema, cerebral oedema, seizures, coma and death. The operative risks include perforation, haemorrhage, air embolism, bowel or bladder injury and cervical trauma (Cooper & Brady 2000). Traditional hysteroscopic resection is highly operator-dependent. The skill of both the surgeon and the operating room staff can influence the likelihood of success or failure (Roy & Mattox 2002).

22 2.3.2 Second-generation procedures

Because hysteroscopic methods are relatively difficult to learn and are associated with operative risks, new methods have been developed. Second-generation, non- hysteroscopic techniques are designed to ablate the full thickness of the endometrium by the controlled application of heat, cold, microwave or other forms of energy. They require sophisticated equipment but less specialized surgical skill than the hysteroscopic procedures, rendering them more user- friendly. Systematic comparisons between first- and second-generation techniques suggest that newer techniques have shorter operating times, better success rates and less complications, but more technical difficulties (Garside et al. 2005, Lethaby et al. 2009).

The uterine thermal balloon catheter endometrial ablation system (TBA)

Balloon systems use an inflatable balloon at the end of a disposable probe, which is inserted into the uterine cavity and through which heated fluid is circulated. When inflated, the balloon conforms to the shape of the cavity, ensuring contact between the heat source and the endometrium. The balloon is filled until the intrauterine pressure reaches 160 to 180 mmHg. Inability to obtain suitable pressure may indicate a uterine perforation or a uterine cavity too large for the balloon. Temperatures vary from 75 to 87 °C and treatment time is 8–15 min. If the pressure or temperature deviates outside prescribed parameters, the system automatically shuts down. If a uterine perforation occurs, the pressure decreases and deactivates the system. There are several devices in the market: ThermaChoice TM (Gynecare Division of Ethicon Inc., Somerville, NJ, USA), the oldest of the second-generation systems, which was approved for marketing in 1997. It is composed of an electronic controller, a connection cord, and a disposable 5-mm catheter that houses the silicone balloon, heating element, and thermocouples. The controller continuously displays the intrauterine pressure, the temperature of the fluid, and the treatment time. ThermaChoice® uses sterile 5% dextrose in water, which is circulated in a closed-loop system. Cavaterm TM (Pnn Medical SA, Morges Switzerland) and MenoTreat TM (Atos medical Horby, Sweden) heat the saline or glycine within a separate unit (Munro 2004).

23 Fig. 3. Thermal balloon endometrial ablation device (Cavaterm®). Picture from the manufacturer.

Hydrothermoablation

Hydro ThermAblator Endometrial Ablation System (HTA; BEI Medical Systems, Inc., A Boston Scientific Company, Teterboro, NJ) was approved for marketing in 2001. This is a technique where externally heated saline is infused to the endometrial cavity through the inflow channel of an operating 3-mm hysteroscope in a disposable 7.8-mm sheath. It is not expected that fluid will pass through the fallopian tubes at pressures under 70 mmHg. Treatment time is 10 min at an intrauterine temperature of 90 °C. A potential advantage of free fluid is the ability to perform the operation in the presence of congenital malformations or leiomyomas. Unlike other second-generation techniques, the procedure is performed under direct hysteroscopic visualization. A hydrothermoablation device monitors fluid loss and switches off if fluid is lost from the uterine cavity (Sowter 2003).

24 Radiofrequency (RF) electrosurgery

This technique uses alternating current and either monopolar or bipolar electrodes to ablate the endometrium. VestaBlate (Vesta Medical, Mountain View CA, USA) consists of a silicone inflatable balloon covered with 12 monopolar electrodes and a controller to monitor and distribute current from a matched electrosurgical generator. Electrode temperature is maintained at 75 °C for 5 min. NovaSure TM system (Hologic, Marlborou, USA) was approved for marketing in 2001 and requires the shortest treatment time (90 seconds) of the new ablation technologies. It is suitable for uterine cavities up to 6–10 cm in length. The device uses bipolar energy dispersed by a disposable conductive copper mesh that expands in a triangular shape to conform to the endometrial cavity. The cornu-to- cornu width is measured by the device and this measurement, along with the endometrial cavity length, is entered into the radiofrequency generator. A cavity perforation test is performed by inflating the endometrial cavity with CO2. Intrauterine pressure must be maintained for 4 seconds for a successful test. The procedure requires dilatation to 8–8.5 mm, and the treatment temperature is 90 ºC. A vacuum draws the endometrium into better contact with the mesh (Roy & Mattox 2002).

Fig. 4. Radiofrequency endometrial ablation device (Novasure®). Picture from the manufacturer.

25 Microwave endometrial ablation

MEA (Microsulis Americas Inc, Boca Raton) uses the application of electromagnetic energy to the endometrium. With a microwave frequency of 9·2 GHz, heat penetrates to a depth of 6 mm from the hemispherical tip of an intrauterine probe. Temperature is displayed and the operator ensures that the temperature of the probe tip remains within 75–80 °C as the probe is moved slowly from side to side and gradually withdrawn. Treatment time is 3–6 min depending on cavity size, and uterine cavities up to 14 cm can be treated (Carter 2005). More recently, this ablation system has been withdrawn from the European market.

Laser ablation

ELITT (endometrial laser intrauterine thermotherapy) uses diffused laser energy from an intrauterine array of three 830 nm diode-laser fibres. Two lateral fibres and a single central fibre on a disposable handset are introduced into the uterus and the cavity is illuminated for 7 minutes (Sowter 2003). This device is no longer marketed.

Endometrial cryoablation

Cryoablation was first evaluated over 40 years ago, but consistently effective devices have only recently become available. Her-Option (CryoGen Inc., San Diego, CA)), approved in 2001, uses a disposable 5-mm probe (metallic tip surface temperature -100 °C) that is inserted into the uterus for two freeze-thaw cycles of 5–7 min. Ultrasound is used to monitor the depth of tissue freezing by monitoring the expansion of the hyperechoic leading edge of the “ice ball” created by the extreme cold temperatures. The operator stops the freeze cycle if the ice ball approaches the serosal surface of the uterus (Roy & Mattox 2002).

Vapour-based endometrial ablation

AEGEA vapour-based endometrial ablation system (AEGEA Medical, Inc., Redwood City, CA) is a new endometrial ablation device, which uses

26 hyperthermic water vapour (high quality steam) to fully ablate the endometrial cavity. The vapour is introduced via a fenestrated 6 mm outer diameter cannula placed in the lower uterine segment. The system is designed to achieve and monitor pressures less than 50 mmHg. A balloon is used to set the lower cavity’s ablation limit and protect the endocervix. This system is not yet in clinical use, and further evaluation of the safety and ablation profile of this endometrial ablation system is warranted (Garza-Leal et al. 2011).

Table 1. Current second-generation endometrial ablation systems.

Product Company Technique Min uterine Max uterine Method Treatment Cervical length (cm) length (cm) time dilatation (min) (mm)

ThermaChoice® Gynecare Balloon 4 10 87º C fluid 8 5 Cavaterm® Pnn Medical Balloon 4 10 85º C fluid 15 7 Menotreat® Atos Medical Balloon 4 10 85º C fluid 11 8 HydroThermablation® Boston Hydrothermal 4 8 Circulating 10 8 Scientific saline of 90ºC Heroption® CryoGen Cryotherapy n/a 10 Freezing - 14 5.5 100º C ELITT® Laser Thermal 7 7 energy MEA® Microsulis Microwave 6 14 Magnetic 3.5 8 energy VestaBlate® Vesta Bipolar 6 10 Radio 4 8.5 Medical electrode frequency Novasure® Hologic Bipolar 6 10 Radio 1.5 8.5 electrode frequency

2.3.3 Anaesthesia and analgesia

First-generation hysteroscopic endometrial ablation procedures were performed under general anaesthesia, as have the second-generation procedures generally been. The simplicity of use and short treatment times of second-generation ablation techniques have increased the likelihood of acceptable treatment under local anaesthetic. Once treatment under local anaesthesia and possible conscious sedation has become accepted, the next step has been to move the ablation out of the theatre. This has been said to change modern menstrual management dramatically (Cooper 2005). The setting in procedures that are performed under general anaesthesia is usually a daycase setting, in which women, fasted for at

27 least 6 h, are admitted to hospital on the day of the procedure. The procedure is carried out in an operating theatre, usually under propofol anaesthesia with spontaneous breathing. After the procedure the patients stay in a recovery room for a while, and are discharged home from the gynaecological ward in the afternoon or in the evening. Some women are admitted to hospital overnight, for analgesia or due to vomiting. For an outpatient procedure, women take oral analgesics (usually 800 mg ibuprofen and 1 g paracetamol) 2 hours before coming to the ambulatory gynaecological clinic. The anti-inflammatory drug reduces the symptoms of cramping, which is caused by the release of prostaglandins due to cervical manipulation and endometrial ablation. The treatment is given under a paracervical block (e.g. 20 ml ropivacaine 2 mg/ml or 20 ml 3% dental octapressin) and if necessary, alfentanil hydrochloride 0.25–0.5 mg i.v.. Intracervical block with 4% prilocaine has also been successfully used (Prasad & Powell 2008). Afterwards the women recover in a daycase bed and are allowed home after a minimum stay of 2 h. It is essential that the patient is informed in advance of what to expect, and most importantly, a supportive member of staff should engage the patient in distractive conversation throughout the procedure (Cooper 2005). Outpatient procedure under local anaesthesia has been shown to be well tolerated with average maximum VAS scores of 4–6 (Andersson & Mints 2007, Chapa 2008, Clark & Gupta 2004, Penninx et al. 2009, Varma et al. 2010). In a randomized trial (n=191) comparing the acceptability of microwave endometrial ablation using a local anaesthesia/sedation regime or general anaesthesia, the anaesthetic type made no significant difference to the proportion of women describing treatment as totally or generally acceptable (Wallage et al. 2003). Even outpatient procedure without local anaesthesia has been described to be successful (Marsh et al. 2005, Marsh et al. 2007), and as many as 70% of women preferred to have that rather than general anaesthesia. These women wanted to spend less time in hospital and did not want to feel nauseous or ill because of general anaesthesia (Marsh et al. 2008). It seems that due to shorter treatment time, bipolar radiofrequency endometrial ablation may be a slightly more acceptable procedure in office setting than balloon ablation. The postoperative pain also seems to be lower in radiofrequency ablation (Clark et al. 2011). That may be because the burnt endometrium sticks to the copper mesh material of the bipolar radiofrequency device, thus diminishing the amount of prostaglandins.

28 2.4 Macroscopic and histopathologic changes after endometrial ablation

Hysteroscopic appearance of the uterine cavity after endometrial ablation varies considerably. In a study where 22 women were examined with outpatient diagnostic hysteroscopy a median of 11 months (6–16) after thermal balloon ablation, 10 women (45.5%) had a normal uterine cavity, 8 women (36.4%) had intrauterine adhesions, two of whom (9%) had complete obliteration of the cavity (Leung et al. 2003). In a series of 53 patients, outpatient diagnostic hysteroscopy was performed a median of 47 (1–96) months after microwave endometrial ablation. Twenty-five patients (47.2%) had no intrauterine adhesions, while 28 patients (52.8%) showed various forms and extends of intrauterine adhesions on hysteroscopy, five (7.5%) of them having marked adhesions at the level of the internal os, preventing further insertion of the hysteroscopy to assess the uterine cavity. The six patients who had hysteroscopy one month after the ablation had no adhesions at all, and all the obliterations were diagnosed in patients who had had the ablation at least four years earlier. The menstrual outcome was correlated with postablation uterine cavity appearance (Luo et al. 2010). Histological examination of endometrial biopsies or hysterectomy specimens obtained after endometrial ablation show that the thermal injury of the endometrium causes an acute inflammation and necrosis, which can last up to 3 months (Mishra et al. 2003), followed by a chronic repair and regeneration phase with granulomatous and foreign-body giant cell reaction and fibrosis (Colgan et al. 1999, Reid et al. 1992, Tresserra et al. 1999). This often results in striking endometrial scarring and fibrosis even 6 months after ablation (Luo et al. 2010). Long-term histopathologic examination reveals diminished endometrial glands with varied necrosis and scarring, and the number of endometrial glands does not correlate with menstrual pattern (Onoglu et al. 2007, Taskin et al. 2002). A 3- to 6-mm thick, hyalinized, subendometrial band-like zone can be seen even on macroscopic examination of the slides (Fadare et al. 2005). This area is detectable in magnetic resonance imaging as a low signal intensity band, and it corresponds to the region of tissue necrosis (Olson et al. 2002). Sonohysterography reveals a clear-limited hyperechogenic zone surrounding the uterine cavity (Jarvela et al. 2002). Persistent endometrium or endometrial regrowth after endometrial ablation is evidenced by clinical, imaging, and histologic studies (Jarvela et al. 2002, Olson et al. 2002, Onoglu et al. 2007, Taskin et al. 2002, Tresserra et al. 1999, Turnbull et al. 1997). Residual endometrium is seen in up to 95% of women, most

29 commonly at the uterine fundus, close to the tubal ostia. Sometimes the lack of communication between islands of residual endometrium with the uterine cavity results in haematometra formation, dilatation and possibly free intraperitoneal fluid (Olson et al. 2002, Turnbull et al. 1997). Even if all the intrauterine endometrium is successfully destroyed, it can still be present in the intramural oviduct. The other source of persistent endometrial cells after an ablation is from deep adenomyosis. It has been found that approximately 20% of patients with menorrhagia have deep adenomyosis, and this seems to be a major cause of ablation failure (McCausland & McCausland 2007). Tressera et al. found that in 12 subsequent hysterectomies required a mean of 19 months after endometrial loop and rollerball ablation (n=350), there was endometrial regrowth in seven patients, and six of them had adenomyosis. Scarring with formation of additional endometrial cavities was seen in five and endocervical stenosis in two cases (Tresserra et al. 1999).

2.5 Indications and contraindications

Endometrial ablation may be considered in premenopausal women for the treatment of HMB when medical treatments fail, are contraindicated, or are poorly tolerated (Practice Committee of American Society for Reproductive Medicine 2008). In general endometrial ablation can be used to treat HMB for women who have a uterus that is not greatly enlarged (uterine cavity length no more than 11 cm, or 14 cm in MEA) and does not contain large fibroids that distort the uterine cavity. The regularity of the endometrial cavity should be ensured with saline hysterosonography, and possible endometrial polyps have been recommended to be removed before ablation. An endometrial biopsy should be taken to exclude endometrial hyperplasy and malignancy. Endometrial ablation is not indicated in postmenopausal women, or in premenopausal women who wish to preserve their fertility. Hardwick and Owen evaluated the adherence of clinicians to the recommendation to treat women with HMB medically in the first instance. The majority (86%) of women undergoing a second-generation endometrial ablation technique received at least one effective medical therapy before any surgical intervention, indicating a high level of compliance with published guidelines (Hardwick & Owen 2004). The most important reasons for women to choose endometrial ablation over hysterectomy have been shown to be the avoidance of major surgery, the rapid

30 return to normal functioning, and the short hospitalization. These women did not consider amenorrhoea to be an important therapeutic endpoint and were prepared for the possible treatment failure and the need for further surgery (Nagele et al. 1998). In another study, 60% of patients who opted for hysterectomy stated that they would have preferred a noninvasive treatment if the success rate of this type of treatment were >80% (Bourdrez et al. 2004). Endometrial ablation is also a safe treatment option for HMB in high-risk surgical candidates (Aletebi et al. 1999).

2.5.1 Endometrial ablation in the presence of myomas or adenomyosis

Myomas are usually considered as a contraindication for an endometrial ablation. Submucosal myomas are recognized as a cause of abnormal uterine bleeding, but the role of intramural and subserosal myomas causing HMB is less clear. The HydroThermAblator System circulates free fluid and thus adapts to an irregular cavity. It can be successfully used in women with submucosal myomas up to 4 cm in diameter (Glasser & Zimmerman 2003). The success rate with submucosal myomas seems to be lower than in women with normal cavities (failure rate 23.2% and 3.7%, respectively), but hysterectomy because of abnormal bleeding related to myomas has been avoided in up to 88% in women with myomas, which can be considered as a good result (Glasser et al. 2009). Also bipolar radiofrequency endometrial ablation seems to be effective (success rate 95% in 12 months follow-up) in women with intracavitary disease (submucosal myoma or ) up to 3 cm (Sabbah & Desaulniers 2006). Intramural and subserosal myomas do not appear to decrease the effectiveness of endometrial ablation in controlling bleeding. They may, however, cause problems by their continued growth (Loffer 2006). A rare but potential complication of endometrial ablation in the presence of myomas is myoma necrosis, which may manifest with a delay (Appiah-Sakyi et al. 2004, Goldberg et al. 2005, Robson & Devine 2007). Preutthipan and Herabuta published a study stating that hysteroscopic rollerball endometrial ablation is an effective treatment (success rate 86.8%) of HMB and dysmenorrhea in patients with suspected adenomyosis (Preutthipan & Herabutya 2010). However, the diagnosis of adenomyosis is histologic, so the results of the study remain controversial. In general, the coexistence of adenomyosis with other menorrhagia-causing conditions hampers the success of endometrial ablation, because the disease penetration is usually deeper than the

31 effect of endometrial ablation (Levgur 2007). Fadare et al. described a case where a patient with adenomyosis and intramural myomas had hysterectomy 38 days after radiofrequency impedance controlled endometrial ablation, and a histological examination showed that adenomyotic aggregates and leiomyomata in the myometrium were unaffected by this treatment (Fadare et al. 2005). Endometrial ablation is not contraindicated for cases where adenomyosis is likely, but such cases ought to be followed up, so that diagnosis can be reviewed or revised if symptoms persist. Persistent symptoms will reinforce the likelihood of adenomyosis and an offer of hysterectomy can be made without any further delay (Basak & Saha 2009).

2.6 Complications

2.6.1 Intraoperative complications

Manual resectoscopic endometrial ablation methods were highly dependent on operative skills and had intraoperative complication rates of 0.77% to 1.51% (Overton et al. 1997). Second-generation endometrial ablation devices seem to be more safe, and intraoperative complication rate has been shown to be 0.24% (5,859 procedures estimated). The most common intraoperative complication is uterine perforation (0.17), more rare complications are haemorrhage (0.03%), visceral perforation (0.02%) and cardiovascular/respiratory incident (0.02%). Emergency surgery has been required in 0.03% of patients (Rogerson & Duffy 2002). US Food and Drug Administration Manufacturer and User Facility Device Experience (MAUDE) is a database which consists of all voluntary reports since 1993, user facility reports since 1991, distributor reports since 1993, and manufacturers reports since August 1996. This database has yielded reports of the following major complications until 2003: eight cases of thermal bowel injury (one patient died), 12 cases of uterine perforation in which emergent laparotomy was required, three intensive care unit admissions, one necrotizing fasciitis which eventually lead to vulvectomy, ureterocutaneus ostomy, and bilateral below-the- knee amputations. However, there appear to be issues associated with these complications: overriding the inherent safety mechanism of the instrument (NovaSure®) that is specifically designed to prevent complications (the new NovaSure® devices no longer have an override option), using the device after a known uterine perforation, not using a tight seal to the cervix while using

32 HydroThermAblator®, and most often, using a device in women with large or small uteri (out of indicated size range of the device). It is also impossible to estimate the relative or absolute risk of these complications because this database lacks the information of the number of all procedures performed (Gurtcheff & Sharp 2003). Microwave endometrial ablation device is the only second- generation device reported to have bowel perforation complications (Iliodromiti & Murage 2011, Jamieson et al. 2002).

2.6.2 Postoperative complications

Postoperative complications include the following: infection, pain-related obstructed menses (, postablation tubal sterilization syndrome), failure to control menses (repeat ablation, hysterectomy), risk from pre-existing conditions (endometrial neoplasia, caesarean section) and pregnancy. Most of these complications are a result of the residual endometrium (Sharp 2012).

Infection

In a meta-analysis comparing endometrial ablation procedures, the incidence of infectious complications included the following: (1.4–2.0%); myometritis (0–0.9%); pelvic inflammatory disease (1.1%); and pelvic abscess (0–1.1%) (Lethaby et al. 2000). The symptoms of postoperative infection have typically occurred 3 days after endometrial ablation: fever, uterine or adnexal pain and (Amin-Hanjani & Good 1995, Das et al. 2005, Salmeen & Morgan 2009), but the infection may present even 50 days after endometrial ablation (Roth & Rivlin 2004, Schlumbrecht et al. 2007). Among 8 case reports of significant infection, surgery was performed in all cases including drainage of , drainage of abscess or hysterectomy (5 out of 8), and most patients recovered rapidly, except for one woman with acute renal failure, and a woman with sepsis who subsequently died. Although endometrial ablation is a minimally invasive and quick procedure, like dilatation and curettage or diagnostic hysteroscopy, it results in endometrial destruction and necrosis, which raises further questions about antibiotic prophylaxis (Sharp 2012). However, there is no evidence that prophylactic antibiotics are of value in endometrial ablation (Bhattacharya et al. 1995). The authors of a Cochrane review of prophylactic antibiotics for intrauterine procedures concluded that prophylactic antibiotics might be considered when the risk of infection is high (Thinkhamrop et al. 2007).

33 Pain-related obstructive menses

There are some unique long-term complications after total endometrial ablation. One such problem occurs after the endometrium is destroyed, at which time the myometrium is exposed. When the distension medium is removed, the intrauterine walls collapse on each other (McCausland & McCausland 2002). The necrosis and inflammatory reaction that occurs from thermal damage is a catalyst for scarring to develop between the exposed myometrial walls. This causes an intrauterine contracture or shrinkage, which narrows the cavity and may obstruct the corneal areas if the walls grow together. This is clinically irrelevant unless persistent viable or regenerating endometrium is trapped behind the scar (McCausland & McCausland 2007). Contracture and scarring in the presence of persistent endometrium can result in obstructed menses in the form of central or cornual hematometra, postablation tubal sterilization syndrome (PATSS), retrograde menstruation, and potential delay in the diagnosis of endometrial cancer (McCausland & McCausland 2002). Central hematometra, which has an incidence of 1–3%, is most likely to occur when the cervical canal is damaged at the time of endometrial ablation. Menses are obstructed at the level of the cervix, and therefore, the patient typically presents with cyclic pain. Central hematometra can be seen on imaging (ultrasound or MRI) at the time of cyclic pain. Central hematometra can usually be successfully treated with cervical dilatation (Hill 1994, Sharp 2012). Postablation tubal sterilization syndrome was initially reported in 1993 by Townsend et al. They presented 6 patients with a history of a tubal ligation in whom symptoms of cyclic developed associated with severe unilateral or bilateral lower abdominal pain 6 to 10 months after a rollerball endometrial ablation. The patients were noted to have endometrial cavity scarring with one or both swollen proximal fallopian tubes and/or a corneal hematometra (CH) (McCausland & McCausland 2007, Sharp 2012, Townsend et al. 1993). The incidence of diagnosed CH and PATSS is said to be approximately 6–10% and they usually develop 2–3 years after endometrial ablation. The incidence is probably understated, because diagnosing these conditions can be difficult and often missed (Mall et al. 2002, McCausland & McCausland 2002, McCausland & McCausland 2007). The mechanism of pain in PATSS is thought to be retrograde menstruation of corneal hematometra against an obstructed fallopian tube, causing visceral distension. The definitive treatment of CH and PATSS is hysterectomy with salpingectomy, and the pathologist should be asked to serially

34 section the cornu and proximal tubal segment. Otherwise, the diagnosis of CH and PATSS may be missed, because sometimes the proximal tube grossly is only minimally swollen and does not appear to be a until it is opened and old blood is expressed. The hematosalpinx fluctuates with the menstrual cycle (McCausland & McCausland 2002). Preventing CH and PATSS is challenging. Causland et al. found that partial rollerball endometrial ablation successfully treated menorrhagia but did not cause intrauterine adhesions (McCausland & McCausland 1999). However, this is not possible with second generation, non-hysteroscopic global endometrial ablation devices. It has been suggested that the cornual area should be adequately ablated. However, the balloon system was specifically designed to enter the cornual areas only partially to avoid potential thermal injury to the adjacent bowel (Leung & Yuen 2006). A fan-shaped mesh electrode endometrial ablation device (Novasure®) has been designed to treat also the corneal areas. From 2001 until March 2012, four cases of PATSS-like symptoms have been reported in the MAUDE (Manufacturer and User Facility Device Experience) database (US Department of Health and Human Services, US Food and Drug Administration 2011), but none were available in print in any medical journal. The most worrisome problem related to obstruction of endometrial tissue is the potential delay in diagnosis of endometrial cancer. This issue will be discussed later in chapter 2.11.

2.7 Endometrial ablation and fertility

Endometrial ablation is not contraceptive. The incidence of pregnancy after endometrial ablation has been reported to be as low as 0.7%. However, the rate may be significantly higher in younger ovulatory patients. No series has so far reported the risk of pregnancy occurring after nonhysteroscopic endometrial ablation, but it has been assumed that the risk of pregnancy is higher (estimated at 5%) with nonhysteroscopic techniques, because the destruction of endometrium is more homogenous, which protects the myometrium and reduces the likelihood of adhesions (Gervaise et al. 2005). Until 2006, 74 pregnancies were reported in the English literature: 2 (2.7%) were extrauterine pregnancies, 28 (37.8%) were terminated, 14 (18.9%) miscarried, 15 (20.3%) delivered prematurely and 15 (20.3%) progressed to term. Overall in pregnancies after endometrial destruction, there is a significant risk of major obstetric complications: antepartum haemorrhage, intrauterine growth restriction, premature rupture of membranes,

35 and preterm delivery (Lo & Pickersgill 2006). Placenta accreta (Hoffman & Sciscione 2004, Kuzel et al. 2010), uterine rupture (Bowling & Ramsey 2010), fetal malformations due to uterine synechiae (Mukul & Linn 2005) and cervical (Giarenis et al. 2008) have also been described.

2.7.1 Endometrial ablation after previous caesarean delivery

The American Congress of Obstetricians and Gynecologists has recently expressed concern about performing endometrial ablation in cases in which the myometrium can be thin, specifically after abdominal or laparoscopic myomectomy or caesarean delivery (Khan et al. 2011). Vesico-uterine fistula has been reported after endometrial ablation in a woman with three prior caesarean deliveries (Rooney & Cholhan 2010). In addition to that, the MAUDE database describes one perforation and thermal bowel injury leading eventually to death in an obese woman who had previously undergone caesarean delivery, and ureteral injury resulting in an ureterocutaneus ostomy without details of the number of prior caesarean sections (Gurtcheff & Sharp 2003). It is hypothesized that cases of asymptomatic lower-segment myometrial dehiscence covered only by a thin serosal layer may be a predisposing risk factor for urinary tract injury (Sharp 2012). In a cohort study of 162 patients with one or more prior caesarean delivery undergoing endometrial ablation with balloon or radiofrequency device, there was no statistically significant difference in complication rates compared to the control group (n=542) (Khan et al. 2011). In a descriptive safety and efficacy study of 116 women including 26 women with previous caesarean section, there were no complications in women with caesarean delivery (Gangadharan et al. 2010). Thermal balloon ablation raises intrauterine pressure, and is contraindicated after uterine surgery such as classic caesarean section, transmural myomectomy, or when the uterine wall at the scar is less than 8 mm thick. In theory, it might be argued that in these patients, one of the other global ablation systems that do not involve the intrauterine pressures associated with balloon ablation, such as hydrothermoablation or radiofrequency ablation, might be safer (Gangadharan et al. 2010).

2.7.2 Concomitant hysteroscopic sterilization

Permanent sterilization is desired by many of women who choose endometrial ablation. Some investigators recommend systematic laparoscopic tubal ligation

36 for contraception after endometrial ablation, but the supplemental morbidity of a laparoscopy only for tubal ligation seems excessive in relation to the risk of pregnancy (Gervaise et al. 2005). Hysteroscopic sterilization offers a transvaginal approach similar to the endometrial ablation techniques. This has the advantage of being performed at the same time and avoiding abdominal incision with its inherent risks. Valle et al. evaluated the safety of combining transcervical sterilization with intratubal insert and endometrial thermal balloon ablation in a study (n=9) where both procedures were performed just before hysterectomy required for benign uterine bleeding. To measure any temperature increase, needle-style thermocouples were placed along the fallopian tubes and the uterine walls of patients who were undergoing abdominal hysterectomy. Unilateral Essure® microinserts were placed, followed by endometrial balloon ablation and hysterectomy. No disturbance of the intratubal insert devices was noted, and there was no conduction of heat to the surrounding tissue (Valle et al. 2006). Donnadieu et al. showed that concomitant hysteroscopic tubal sterilization can be performed either before or after endometrial ablation, although visualization of tubal ostia poses a greater challenge after endometrial ablation (Donnadieu et al. 2007). Mircea et al. published a study of 100 women undergoing concomitant radiofrequency endometrial ablation and hysteroscopic sterilization in an outpatient setting using conscious sedation. The combined procedure appeared to be safe and feasible, and the mean duration of the procedure was on average only 17 minutes. A small proportion of patients developed post-procedure intrauterine synechiae that impaired the ability to document tubal occlusion by hysterosalpingogram (Mircea et al. 2011), but tubal occlusion can also be documented with ultrasound. In a study of 150 women with menorrhagia and the need of contraception, women were offered combined treatment with simultaneous endometrial ablation and the levonorgestrel-releasing intrauterine device. 105 (70%) women returned a structured questionnaire mailed after 6–54 months (mean 25 months). 51% described their periods as being lighter than normal and 46% had become amenorrhoeic. Overall, 98% stated that they were very satisfied with the treatment (Vaughan & Byrne 2012).

37 2.8 Outcome

2.8.1 Reduction of bleeding

The results of first-generation endometrial ablation techniques are left out of this review, because they are no longer the principal methods since the development of non-hysteroscopic global endometrial ablation devices (Munro 2006). After balloon endometrial ablation the amenorrhea rates in 2- to 6-year follow-up studies are 14–58%, and the combined amenorrhea and hypomenorrhoea rates (=success rates) 68–91% (Amso et al. 2003, El-Toukhy et al. 2004, Mettler 2002, Munro 2004, Sambrook et al. 2010). Differences in the rates may be due to differences in patient selection in some retrospective studies. At a follow-up of 5 and more years, a rise in amenorrhea rate may be attributable to the aging of the patient population enrolled in the study and their transition to the menopause, resulting in physiologic amenorrhea (Gallinat 2007). A summary of long-term studies with a follow-up time of 4 years or more can be seen in Table 2. The amenorrhea rates of impedance-controlled endometrial ablation vary from 46% to 75%, and the amenorrhea rate has a trend of getting better during time (Baskett et al. 2005, Cooper et al. 2002, Fulop et al. 2007, Gallinat & Nugent 2002, Gallinat 2004, Gallinat 2007). A more detailed comparison between different second-generation ablation devices will be presented in chapter 2.8.5. Endometrial ablation seems to be an effective treatment choice for women with congenital or acquired presenting with menorrhagia. El-Nashar et al. found no significant difference in results between 41 women with a coagulopathy and a reference group without bleeding disorders. (El-Nashar et al. 2007, El-Nashar et al. 2010)

2.8.2 Probability of and reasons for hysterectomy

One aim of EA is avoiding hysterectomy. Hysterectomy rates in studies with 4 to 10 years of follow-up vary from 2.9% to 22% (Amso et al. 2003, Comino & Torrejon 2004, Fulop et al. 2007, Furst et al. 2007, Gallinat 2007, Kopeika et al. 2011, Longinotti et al. 2008). In most studies the majority of hysterectomies have been performed during the first 2 years following EA (Comino & Torrejon 2004, Fulop et al. 2007, Gallinat 2007) or at least during 4 to 5 years after EA (Amso et al. 2003, Furst et al. 2007). Menorrhagia was reported as the main reason for hysterectomy in one study (Amso et al. 2003). Other studies report metrorrhagia,

38 fibroids, pelvic/abdominal pain or dysmenorrhea as reasons for hysterectomy that are at least as common as menorrhagia (Comino & Torrejon 2004, Fulop et al. 2007, Furst et al. 2007, Gallinat 2007). The largest study of risk factors for hysterectomy after endometrial ablation is a retrospective cohort analysis of 3,681 women who had undergone EA 3–18 years earlier. Cox regression analysis found that age was the only predictive factor for hysterectomy. Women aged 45 years or younger were 2.1 times more likely to have hysterectomy than women aged older than 45 years. Hysterectomy risk increased with decreasing age and exceeded 40% in women aged 40 years or younger. Overall, type of endometrial ablation procedure, setting of endometrial ablation procedure, and presence of leiomyomas were not predictors of hysterectomy (Longinotti et al. 2008).

Table 2. Table Long-term outcome studies with different second-generation ablation devices.

Author, Year Device Number of Follow-up Amenorrhoea Amenorrhoea+ Hysterectomy women time (%) eumenorrhoea (%) (years) (%) Amso, 2003 Balloon 188 4-6 471 90,61 13.3 Mettler, Balloon 60 4 58 100 0 2002 Sambrook, Microwave 154 5 841 981 11 2010 Kopeika, Hydrothermal 376 8 38 75 11 2011 Gallinat, Radiofrequency 107 5 75 98 2.9 2007 Fulop, 2007 Radiofrequency 75 7 97 100 8 1 of the non-hysterectomised

There are very scarce data of the uterine pathologic features in women who have undergone hysterectomy because of failed EA. In a series of 67 patients rates of pathologic findings in hysterectomy specimens after failed EA were determined. The reasons for hysterectomy in the 67 patients with available pathology reports were bleeding in 34 (51%), pain in 19 (28%), and bleeding and pain in 14 cases (21%). The pathology reports of these specimens showed leiomyomas in 33 specimens (49%); intramural myomas were present in 15 women (44%) who underwent hysterectomy because of bleeding and in 8 women (42%) who underwent hysterectomy because of pain. Hematometra was identified in 7 pathologic specimens (10%). Specifically, hematometra was identified in

39 specimens from 5 out of 19 women who underwent hysterectomy because of pain. As conclusion, hematometra was a significant finding in women who underwent hysterectomy because of persistent pain after EA. Intramural leiomyomas may have been a possible pathologic predictor of EA failure. Patients were categorized into subgroups on the basis of the clinical reason for hysterectomy, and there were no significant differences in findings between subgroups (Table 4) (Carey et al. 2011).

2.8.3 Prognostic factors

The largest study evaluating the prediction of treatment outcomes after global endometrial ablation is by El-Nashar et al. From January 1998 through December 2005, 816 women underwent global endometrial ablation with either a thermal balloon ablation or radio frequency ablation device. The risk factors for treatment failure (defined as hysterectomy or reablation because of bleeding or pain) included age younger than 45 years, parity of 5 or greater, prior tubal sterilization, and history of dysmenorrhea. The device, BMI, bleeding profile, size or position of the uterus, myomas, polyps or endometrial thickness were not predictive factors for treatment failure (El-Nashar et al. 2009). Age, previous tubal ligation and dysmenorrhea have also been shown to be predictive factors in other studies (Bongers et al. 2002, Gemer et al. 2003, Longinotti et al. 2008, Mall et al. 2002). There is some evidence that hysterectomy risk is higher for sterilized than for nonsterilized women in overall female population, and it has been speculated that sterilized women may have increased willingness to undergo other elective surgical procedures (Hillis et al. 1998). Another reason may be the development of PATTS. The relationship of preoperative endometrial thickness and clinical outcome of EA is controversial: in some studies it has appeared to be a predictive factor (Bongers et al. 2002, Istre et al. 1996a), but mostly this has not been the case (El-Nashar et al. 2009b, Lok et al. 2003, Shaamash & Sayed 2004). Endometrial thinning with GnRH analogues, danazol or progestogens seems to improve both the operating conditions for the surgeon and short-term post-operative outcome in first-generation hysteroscopic endometrial ablation procedures (Bhatia et al. 2008, Shamonki et al. 2000, Sowter et al. 2002), but this effect has not been shown with second-generation procedures (Jack et al. 2005, Sowter et al. 2002). Maintaining high intrauterine pressure during the treatment cycle with an endometrial thermal balloon ablation device can be a predictive factor for successful treatment (Lok et al. 2003,

40 Shaamash & Sayed 2004). Body mass index (BMI) does not seem to be a predictive factor until it reaches 34 or more (El-Nashar et al. 2009b, Fakih et al. 2011).

2.8.4 Patient satisfaction

There is emerging evidence that improvement in quality of life is more relevant to women than amenorrhea rates (Amso 2006). Despite this, only few studies have focused on patient satisfaction. Satisfaction rates after balloon and microwave endometrial ablation have been 71–83%, which are higher than amenorrhea rates (El-Toukhy et al. 2004, Furst et al. 2007, Jarrell & Olsen 2003, Mall et al. 2002, Sambrook et al. 2010). The patient satisfaction rate after radiofrequency endometrial ablation has been shown to be up to 84–90% (Elmardi et al. 2009, Bongers et al. 2005). By identifying predictors of treatment outcomes after EA, patient counselling could improve and failure rates could decrease with better patient selection. Some women may expect amenorrhea after treatment, and knowing the patient-specific likelihood of amenorrhea optimizes preoperative counselling. This information may improve patient satisfaction and may reduce the need for additional intervention after ablation (El-Nashar et al. 2009).

2.8.5 Comparison of different second generation techniques

The rapid development of a number of new methods of endometrial destruction has made systematic comparisons between methods difficult (Lethaby et al. 2009). There are, however, numerous studies comparing bipolar radiofrequency endometrial ablation with balloon endometrial ablation. In those studies there were no significant differences in failure rates, but women treated with radiofrequency ablation were more likely to have amenorrhea (Abbott et al. 2003, Bongers et al. 2004). In a five-year follow-up of a randomized controlled trial of 126 women comparing bipolar radiofrequency and balloon endometrial ablation, the total response rate was 96% in the bipolar group and 90% in the balloon group. Amenorrhea was reported in the bipolar group by 48% and in the balloon arm by 32% of the women. Hysterectomy rates were 9.8% in the bipolar group and 12.9% in the balloon group (Kleijn et al. 2008). In a population-based cohort of 455 women (255 with radiofrequency ablation and 200 with balloon ablation), women who underwent radiofrequency ablation were 3 times more likely to have postprocedure amenorrhea (El-Nashar et al. 2009a). However, despite better

41 amenorrhea rates after radiofrequency ablation, there seems to be no difference in health-related quality of life between women treated with radiofrequency ablation and women treated with balloon ablation (Bongers et al. 2005). Patients treated with radiofrequency ablation have reported statistically significantly lower intraoperative and postoperative pain than those treated with balloon ablation. At least three factors may explain this difference. First, distension of the uterine cavity can generate pain during and after the procedure. The balloon catheter applies a distending pressure of 160 to 180 mmHg to the uterine wall, whereas in the radiofrequency system suction is employed to engage endometrial tissue in permanent and close apposition to the electrode. Second, time of tissue exposure to the destructive element (heat) is longer with the balloon device compared to the radiofrequency system. Finally, by-products of ablation may penetrate into myometrium and blood stream. Some biologically active substances released during tissue death (prostaglandin-F2α) may interact with uterine muscle, causing cramping, and because they are lipid soluble they can pass easily through cell membranes. With the balloon system, ablation by- products are continuously directed into tissue by a pressurized balloon, which acts as an impermeable membrane. With the radiofrequency system, these by-products are continuously evacuated from the uterine cavity by suction (Laberge et al. 2003). The Food and Drug Administration (FDA) has provided an overview of the following second-generation endometrial ablation techniques: Thermachoice®, HydroThermablator®, Her Option® and Novasure®. It is based on Summary of Safety and Effectiveness Data (SSED) documents, which detail aspects of safety, efficacy, performance and other clinical use issues of importance. According to those data, Thermachoice and Novasure had the highest success rates at 12 months, Novasure and Hydrothermablator had the highest amenorrhea rates at 12 months, and Novasure had the lowest adverse event rates in the first 24 hours, between 24 hours and 2 weeks and between 3 weeks and 1 year of follow-up (Cooper & Gimpelson 2004). However, in a series of 160 women with a follow- up of 12 months, bipolar radiofrequency endometrial ablation seemed to be superior to hydrothermablation according to patient satisfaction (87% vs. 68%) and amenorrhea (47% vs. 24%) (Penninx et al. 2010). It seems that there are no differences in results with microwave and balloon ablation (Sambrook et al. 2009). In a postal questionnaire survey on preferences of British consultant gynaecologists on the use of endometrial ablation techniques in 2008, TBA was

42 the preferred method (32.1%), followed by microwave endometrial ablation (29.8%), while the percentage of RFEA was only 9.8% (Deb et al. 2008). Recently RFEA has been the most rapidly rising technique used, and the number of RFEA-procedures is about twice as many as that of TBA-procedures (Daniels et al. 2012). Unfortunately there are no such data available concerning the preferences among Finnish gynaecologists.

2.9 Outcome of endometrial ablation compared to levonorgestrel device

In many countries, levonorgestrel-releasing intrauterine system (LNG-IUS; Mirena, licenced 1990 in Finland) is the first-line pharmaceutical treatment for idiopathic menorrhagia. It delivers LNG 20 µg to the endometrium per 24 h for 5 years, but the amount of daily released hormone decreases over the course of time. Short-term follow-up studies have documented acceptable continuation rates with high tolerability and user satisfaction (Goni et al. 2009, Lete et al. 2008, Robinson et al. 2008), but there is growing evidence suggesting limited satisfaction with LNG-IUS in women because of high incidence of adverse effects leading to the removal of the device before the 5-year effective life-time recommended by the manufacturer (Ewies 2009). A large observational study of 17,360 Finnish women based on a questionnaire with a response rate of 75% reported a 65% continuation rate after 5 years. The study population consisted of all LNG-IUS users, including women who had LNG-IUS for contraception. (Backman et al. 2000). In a randomized trial with 5 years of follow-up, 42% of the women randomized to the LNG-IUS as treatment of menorrhagia eventually underwent hysterectomy (Hurskainen et al. 2004). The reasons of early removal of LNG-IUS are described to be unscheduled bleeding, progestogenic adverse effects (bloating, weight gain, headaches, depression, breast tenderness, excessive hair growth, greasy skin, acne and sexual disinterest). Many women have more than one indication for removal (Daud & Ewies 2008). Unfortunately there are no long-term follow-up studies published in English literature comparing LNG-IUS and endometrial ablation in treatment of menorrhagia. In short 6- to 12-month follow-up both LNG-IUS and thermal balloon ablation (TBA) appeared to be effective in controlling heavy menstrual bleeding, but patients with TBA had less and fewer women from the TBA group seemed to end up with hysterectomy (Barrington et al. 2003, de Souza et al. 2010, Shaw et al. 2007). In two randomized, controlled

43 studies with a follow-up of two years, treatment with the LNG-IUS was associated with lower PBAC scores and a higher amenorrhea rate than thermal balloon ablation, but both treatments were associated with good therapeutic effects and similarly high levels of patient satisfaction and overall quality of life. There was also no difference in the number of women requesting an alternative treatment. The sample size of the studies was too small to adequately assess adverse events. However, nonserious adverse events (mostly hormonal) were more frequently reported in the LNG-IUS group than in the endometrial ablation group (Busfield et al. 2006, Kaunitz et al. 2009).

2.10 Outcome of endometrial ablation compared to hysterectomy

The question of what is the best option for treatment of menorrhagia after medical treatment has failed is a difficult one. Hysterectomy is a definitive treatment of menorrhagia, but it is a major surgical procedure with possible physical and sometimes also emotional complications as well as significant costs. Systematic review and meta-analysis of data from individual patients (n=2,814) sought from randomized trials evaluating patients’ dissatisfaction showed that more women were dissatisfied after second-generation endometrial ablation (11%) or levonorgestrel-releasing intrauterine device (17%) than after hysterectomy (5%). Length of the uterine cavity ≤8 cm and absence of fibroids/polyps were associated with reduced dissatisfaction after endometrial ablation. Hysterectomy was associated with increased length of stay in hospital and a longer recovery period. There were no randomized trials comparing second-generation ablation and hysterectomy, so the comparisons were indirect. In addition, the follow-up was only 12 months (Middleton et al. 2010). A randomized controlled trial of 237 women assigned to either endometrial ablation (hysteroscopic or balloon) or hysterectomy (vaginal, laparoscopic or abdominal) with a follow-up of up to 5 years showed that at 24 months both treatments were effective in dysfunctional bleeding. By 48 months, 32 of the 110 women initially receiving endometrial ablation required reoperation, which can be due to the inclusion criterion of 18 years (Dickersin et al. 2007). More specific was a study comparing thermal balloon ablation (n=34) with laparoscopic supracervical hysterectomy (n=34) in women aged 35–50 years with menorrhagia. Pictorial Blood Loss Assessment Chart (PBAC) score after 24 months was significantly reduced in both groups, but hysterectomy showed a negative impact on the emotional state. However, LSH showed a definitive improvement of the symptoms and a better life quality profile

44 (Sesti et al. 2011). In a population-based comparison of psychosexual health of 8,900 women 5 years after endometrial resection/ablation (n=3,845) or hysterectomy (vaginal, abdominal or laparoscopic) with (n=2,305) or without (n=3,397) bilateral oophorectomy, prevalence of self-reported psychosexual problems was slightly higher in the hysterectomy group even if at least one ovary was conserved (McPherson et al. 2005). A retrospective analysis of hospital episode statistics in Scotland between 1989 and 2006 comparing outcomes following hysterectomy or endometrial ablation assessed the risk of further gynaecological surgery. A total of 37,120 women had a hysterectomy, 11,299 women underwent endometrial ablation without a subsequent hysterectomy and 2,779 women underwent endometrial ablation followed by a subsequent hysterectomy. The median follow-up was 11.6 years in the hysterectomy and 6.2 years in endometrial ablation cohort. Compared with women who underwent hysterectomy, those who underwent ablation were less likely to need pelvic floor repair (adjusted hazards ratio 0.62) or tension free vaginal tape surgery for urinary incontinence (adjusted hazards ratio 0.55). The risk for adnexal surgery was also smaller in women with ablation than in women with hysterectomy (adjusted hazards ratio 0.80) (Cooper et al. 2011).

2.11 Endometrial ablation and endometrial cancer

In theory, the obstruction of the uterine cavity may mask uterine bleeding, which is typically a warning signal for intrauterine neoplasia. The synechiae may also make endometrial sampling difficult, especially from upper fundal and corneal areas. There are several case reports of endometrial cancer diagnosed at the time of or after endometrial ablation. Most patients had risk factors for and cancer and were therefore poor candidates for endometrial ablation (AlHilli et al. 2011, Areia et al. 2006, Brooks-Carter et al. 2000, Cooper & Brady 2000, Le Marrec et al. 2010, Lee et al. 2005, Perez-Medina et al. 2003, Sagiv et al. 2005, Sinervo & Martyn 2000, Valle & Baggish 1998). The systematic literature review identified 22 endometrial cancer cases occurring after EA. Most (76.5%) were stage I at diagnosis. Time to endometrial cancer diagnosis ranged from 2 weeks to 10 years following EA. Contrary to the concerns, it seems that most patients with endometrial cancer after endometrial ablation demonstrate symptoms such as bleeding and . In most cases, a preoperative diagnosis is feasible (AlHilli et al. 2011). Kroogh et al. conducted a follow-up study which considered 367 women 11 years after endometrial ablation.

45 Using registered data on cancer, they observed three women with incidental endometrial cancer and the expected numbers were 6.8 cases (Krogh et al. 2009). This is in line with a retrospective cohort study of 509 women who had undergone hysteroscopic endometrial ablation 7–24 years earlier. A total of 5,063 woman-years were identified with two cases of endometrial cancer. The expected incidence was 1.66 cases in an age-matched group (Neuwirth et al. 2004). Although ablation itself does not appear to increase the risk of subsequent endometrial cancer, it may be recommendable to avoid ablation in women who are at an increased risk of developing endometrial cancer. These risks include obesity, diabetes mellitus, nulliparity, chronic , tamoxifen therapy and hereditary non-polyposis colorectal cancer. Endometrial hyperplasia should also be considered a contraindication to endometrial ablation (McCausland & McCausland 2007, Sharp 2012). In such cases, endometrial ablation should only be recommended under circumstances when the risk of hysterectomy far outweighs treatment by endometrial ablation (Valle & Baggish 1998).

2.12 Hormone replacement therapy after endometrial ablation

There are very little data concerning hormone replacement therapy after endometrial ablation. It is obvious that unopposed oestrogen therapy should not be used after endometrial ablation (Istre et al. 1996b). In a Danish 11-year follow- up of 367 women who had previously had endometrial resection, hormonal treatment did not seem to have any influence on the probability of later hysterectomy. Some 26% of women who had hormonal treatment after endometrial resection (n=77), and 34% of women who did not have hormonal treatment, had a hysterectomy later (Krogh et al. 2009).

2.13 Economics

Determining the cost-effectiveness of a treatment requires calculating direct medical costs (theatre resource use, equipment and drug cost, staff costs, hospital stay), complications, indirect costs from sick leave, estimated effects on health- related quality of life (usually expressed in quality-adjusted life-years, QUALYs) and probability of medical treatment for failed procedures. The calculations can be based on real resource use or various economic modellings, which rely upon previously published studies.

46 Second-generation techniques are in most cases likely to be more cost- effective than first-generation techniques (Garside et al. 2004). There are only few studies comparing the cost-effectiveness of different second-generation endometrial ablation procedures, and none of them include bipolar radiofrequency ablation. According to one study, microwave ablation is likely to be more cost- effective than thermal balloon ablation (Kilonzo et al. 2010), but another study shows the opposite result (Garside et al. 2004). Levonorgestrel-releasing intrauterine system seemed to be more cost- effective than endometrial thermal balloon ablation performed in outpatient setting in a short 2-year follow-up study with a relatively small sample size (79 women) (Brown et al. 2006). Hysterectomy was suggested to be more cost-effective than endometrial ablation in a study using the Markow model. In that study, 52% of ablations were supposed to be performed under general anaesthetic. Both first- and second- generation ablation methods were included, except for bipolar radiofrequency ablation (Garside et al. 2004). Hysterectomy is suggested to produce more QUALYs than endometrial ablation and LNG-IUS, so it is considered the most cost-effective option in many studies, even if it is more costly than endometrial ablation and LNG-IUS. In these studies, the failure rate for endometrial ablation is assumed to be 25–35%, and the type of anaesthesia was not always reported (Bhattacharya et al. 2011, Roberts et al. 2011, You et al. 2006). The lower the probability of need for extra surgery after endometrial ablation is, the more cost- effective the procedure appears (You et al. 2006). An important aspect of the cost-effectiveness of ablation techniques is the outpatient possibility. If the procedure can be performed under local anaesthesia, considerable cost savings can be achieved because an expensive operating theatre with its equipment and trained staff is not necessary (Hurskainen 2006). There is only little difference in cost when treatments are performed under local or general anaesthetic in theatre (Seymour et al. 2003), but taking the procedure out of the theatre and performing it as an outpatient procedure under local anaesthetic results in significantly reduced health service costs (Jack et al. 2005). More research is needed to evaluate the clinical effectiveness and cost-effectiveness of the best second-generation endometrial ablation technique under local anaesthetic versus Mirena and laparoscopic hysterectomy versus second-generation endometrial ablation (Bhattacharya et al. 2011). Clegg et al. estimated (with Markow model) that LNG-IUS followed by ablation, even if it were followed by hysterectomy in case of treatment failure,

47 offers a cost-effective treatment for menorrhagia when compared to immediate surgery (ablation or hysterectomy). In this study the expected outcomes were that 41.7% of patients with LNG-IUS would require surgery within 5 years and that 20.4% of patients with endometrial ablation would undergo hysterectomy within 5 years (Clegg et al. 2007).

48 3 Aims of the study

The purpose of the present study was to evaluate the long-term effects of endometrial ablation in treatment of heavy menstrual bleeding, avoiding hysterectomy, to evaluate patient satisfaction and later endometrial assessment. The possibility of reducing the health economics costs of endometrial ablation was assessed, and finally, an attempt was made to reduce the formation of intrauterine adhesions after endometrial ablation. The specific aims of the study were as follows:

1. To assess the percentage of patients having to undergo a hysterectomy 6 years after endometrial ablation and report the indications for the hysterectomy. Patient satisfaction was also evaluated (I). 2. To evaluate the possibility for outpatient assessment of the endometrium 6 years after endometrial ablation (II). 3. To estimate the cost savings when endometrial ablation with bipolar radiofrequency device is taken out of the operation theatre to be performed as an outpatient procedure under local anaesthetic (III). 4. To evaluate the effect of hyaluronic acid gel in preventing the formation of intrauterine adhesions after endometrial ablation with bipolar radiofrequency device (IV).

49

50 4 Materials and methods

4.1 Study subjects

The data for assessing long-term effects of endometrial ablation (Study I) were collected from the hospital records of Oulu University Hospital and Länsi-Pohja Central Hospital in Kemi. Medical records of 190 women who had undergone endometrial thermal ablation in those hospitals between 1996 and 2003 were reviewed. Women who were postmenopausal at the time of the procedure and women with endometrial hyperplasia were excluded, leaving 172 patients for analysis. Patients’ mean age at the time of the procedure was 42 years (range 32- 52 years). 127 (74%) of women had undergone sterilization before endometrial ablation, and 63 (37%) had tried treatment with intrauterine levonorgestrel- releasing device because of menorrhagia. Among the study population of Study I, 80 women who had not undergone hysterectomy during the follow-up were invited to a clinical examination for evaluation of the feasibility of endometrial assessment (Study II), and a total of 57 women took part in the study. Their mean age was 49 years (range 46–52 years). The time from endometrial ablation was a mean of 6 years (range 3–10 years). The data for cost-minimisation analysis (Study III) was obtained from the hospitals database. The method evaluated was bipolar radiofrequency ablation device Novasure® (Novacept, Palo Alto, California, USA). Detailed data of each endometrial ablation performed under general anaesthesia in the operating theatre during 2009–2010 (n=20) and outpatient procedures performed during October 2010-May 2011 (n=16) were collected. The patients for the evaluation of hyaluronic acid gel in prevention of intrauterine adhesions (Study IV) were recruited among patients referred to the Department of Obstetrics and Gynaecology at Oulu University Hospital for consideration of endometrial ablation between January 2011 and April 2012. Exclusion criteria were abnormal uterine cavity in saline sonohysterography, abnormal histology in endometrial sample and wish for future pregnancies. A total of 32 women with a mean age of 45 years (range 34–52 years) took part in the study.

51 4.2 Methods

4.2.1 Study design

The study on long-term results of endometrial ablation (Study I) was a retrospective study. The data were collected from the medical records of 172 women who had undergone endometrial ablation a mean of 63 months earlier (range 23–110 months). To assess patient satisfaction, a questionnaire was sent to all 190 study subjects. A reminder was sent for missing questionnaires. A total of 167 (88%) subjects returned the questionnaire, and 153 replies were included in the analysis after exclusion criteria. The feasibility of endometrial assessment after endometrial thermal ablation (Study II) was assessed in a prospective observational study. Patients came for clinical examination to the gynaecological outpatient clinic of the Department of Obstetrics and Gynaecology of Oulu University Hospital. All patients gave their informed consent before participating in the study. The economic evaluation method of the cost difference between a daycase endometrial thermal ablation performed under general anaesthesia and an outpatient endometrial ablation using local anaesthetic (Study III) was a cost- minimization analysis based on real reported resource use. The effect of hyaluronic acid gel in prevention of intrauterine adhesions was studied in a randomized, controlled and double blind pilot study (Study IV). The summary of the studies is shown in Table 3. The study protocols of Study II and IV were approved by the Ethics Committee of the Medical Faculty. Studies I and III did not require any interventions, so approval of the Ethics Committee was not needed.

Table 3. The summary of studies.

Study Main outcome(s) Design Number of subjects I Probability of hysterectomy, patient Retrospective observational 172 satisfaction after EA II Feasibility of endometrial Prospective observational 57 assessment after EA III Cost difference between two Cost minimization analysis 36 different EA settings IV Prevention of intrauterine Randomized controlled 32 adhesions after EA double blind

52 4.2.2 Data collection

For the study of long-term effects of endometrial thermal ablation data concerning preoperative anamnesis and status findings, the operation, indication for possible subsequent hysterectomy and the pathologic findings of the removed uterus were obtained. The procedures were performed with different devices, and usually the device was not reported in the records. We know that during that period four different devices were in use, three balloons and occasionally a bipolar radiofrequency ablation device. In the patient satisfaction questionnaire women were asked about postoperative pain, bleeding pattern, dysmenorrhea, climacteric symptoms, use of hormone replacement therapy and whether they would recommend the procedure to their friends. (Study I) When assessing the feasibility of endometrial assessment after endometrial ablation, information concerning the menstrual pattern, pain, climacteric symptoms and the use of HRT was collected. Clinical examination included transvaginal ultrasound, saline sonohysterography and endometrial biopsy with a Pipelle device. Clinical examinations were performed by a single investigator (SA). The samples were sent to histologic analysis performed by a single pathologist. (Study II) For the cost minimization analysis data of each daycase procedure were obtained from the hospital’s operative database, which contains detailed information of every stage of the procedures, including the time spent. Calculations of costs were made by the local hospital financial planning department and they were based on real resource use on average, using 2009 information. Staff costs were calculated based on local salaries for consultant doctors and nurses. The overhead costs were provided by the hospital financial planning department. Detailed assessment of the resource use of each outpatient procedure was made by the nurses responsible for outpatient procedures performed during the study period. Pre-admission consultations and sick leaves were excluded as they were the same for both treatment methods. (Study III) For Study IV, eligible patients who signed a written informed consent were randomly allocated on the day of the endometrial ablation to the study group or to the control group. Randomization was carried out in blocks of four opaque sealed envelopes by a nurse who was responsible for the outpatient procedures on that day. The randomization data were kept apart from patients’ hospital records.

53 4.2.3 Surgical procedures (Study II and IV)

The treatment was given as an outpatient procedure under local anaesthetic (paracervical block with ropivacaine (Naropin® 2 mg/ml total amount of 20 ml)) and intravenous analgesia with 0.5 mg alfentanil (Rapifen®). The patients took 800 mg ibuprofen and 1 g paracetamol before signing up to the clinic. The setting was identical with the outpatient setting in the cost minimization study. Patients allocated to the study group received 10 ml of autocross-linked hyaluronic acid gel (Hyalobarrier gel®; Baxter, Pisa, Italy) applied intrauterinely and patients allocated to the control group had Pipelle device placed in the uterus as a placebo procedure one minute after bipolar radiofrequency endometrial ablation. Outpatient diagnostic hysteroscopy was performed after a mean of 15 weeks. The hysteroscope was Versascope® (Gynecare, Ethicon Inc.), which has a diameter of 3.5 mm and uses saline (NaCl 0.9%) as a distension medium. The patients had been advised to take ibuprofen 800 mg and paracetamol 1 g one hour before the appointment, and no anaesthetics were needed. Anamnestic data were collected concerning bleeding after the procedure, presence of pain and possible postoperative infections. The hysterosopic finding was classified with a classification created for this study. The endometrial ablations and control hysteroscopies were performed by different doctors. All control hysteroscopies were either performed or supervised by the author.

4.2.4 Statistics

Statistical analysis was performed using the SPSS for Windows, versions 14 and 16. χ² (chi-square), Mann-Whitney and Kruskal-Wallis U-tests were used to test the level of significance. A p value <0.05 was considered significant. In Study III statistical analysis was not needed, because the outcome was the cost difference.

54 5 Results

5.1 Long-term effects and patient satisfaction

5.1.1 Avoiding of hysterectomy

Subsequent hysterectomy was performed to 28 (16%) of 172 patients during the long-term follow-up study, meaning that up to 84% of the patients avoided hysterectomy. The hysterectomy was usually performed within a median of 13 months (range 4–106 months) after endometrial ablation. Only 25% of the hysterectomies were performed because of menorrhagia. The indications for hysterectomy were as follows: menorrhagia (7 patients): irregular bleeding (9 patients), pain (5 patients) and fibroids (7 patients). Histologic findings of the removed uterus were as follows: no pathologic findings in 16 patients, uterine fibroids in 9 patients, adenomyosis in 4 patients and endometrial hyperplasia in 2 patients. Women with irregular cycles (n=25) before ablation ended up having hysterectomy more often than women with regular cycles (n=141), 28% vs. 15%, respectively. There was a trend towards women under 40 years having hysterectomy more often than women between 40–45 years and women over 45 years (19% vs. 16% vs. 14%, respectively). Previous sterilization seemed to be a prognostic factor: 19% of the 127 sterilized women and 9% of the 45 non- sterilized women had a hysterectomy during the follow-up. Prior treatment with levonorgestrel-releasing device was linked to higher risk of hysterectomy than not having prior treatment with LNG-IUD, 24% vs. 12%, respectively. None of the variables was statistically significant in this study, which may be due to the rather small sample size, so the differences found may be considered as merely suggestive.

5.1.2 Bleeding profile and dysmenorrhea

According to the 153 questionnaire replies, 60% of women had eumenorrhoea, 16% had amenorrhoea, 24% had persisting menorrhagia and 12% had irregular spotting after the endometrial ablation. Fifty-six per cent of the women with persisting menorrhagia had undergone hysterectomy at some point. Dysmenorrhea was reported by 15 (9.8%) of the women. There were 28 users of

55 hormone replacement therapy amongst the women who returned the questionnaire, and 4 of them reported having irregular bleeding. (Study I) Out of 57 women who came to the clinical examination for the endometrial assessment study (Study II) 65% had eumenorrhoea, 30% had amenorrhoea, 5% still had menorrhagia and 23% had irregular spotting. Dysmenorrhea was reported by 7 (12%) of the 57 women. There were 7 current users of hormone replacement therapy, and none of them had any bleeding problems with the treatment. (Study II) In short term, none of the women (n=27) who came to control hysteroscopy a mean 15 months after bipolar radiofrequency ablation, reported dysmenorrhea. Amenorrhea was reported by 12 women and eumenorrhoea by 14 women. One woman had persistent menorrhagia, and she was diagnosed with submucotic myoma. There were no cases of irregular spotting. (Study IV)

5.1.3 Patient satisfaction

Some 76% of the women who returned the questionnaire (n=167) were satisfied or fairly satisfied, and 77% were willing to recommend the procedure to their friends. (Study I)

5.1.4 Assessment of the endometrium

Endometrial demarcation determined by transvaginal ultrasound was unclear in all 57 women who participated in Study II. Endometrial thickness was dependent on the bleeding profile, being thinnest in amenorrhoeic women (mean 4.5 mm), somewhat thicker in eumenorrhoeic women (mean 5.6 m) and thickest in the three menorrhoeic women (mean 6.6 mm). Outpatient endometrial sampling with Pipelle® device succeeded in 44 (77%) women, and the success was not related to the bleeding profile. The length of the uterine cavity was 0.5–5 cm (mean 2 cm) smaller than before endometrial ablation in 34 (60%) women, the same as before in 4 women and larger than before in 5 women. Total obliteration of the uterine cavity was found in 13 (23%) women. Saline sonohysterography catheter was successfully entered in the uterine cavity a little more often than the Pipelle® device in 47 (82%) women, but the cavity distended normally in only nine (16%) women; in rest of the women the cavity distended either irregularly or was visualized as a narrow tube. Pieces of endometrium were found in histologic analysis of all samples, regardless of the bleeding profile. The cytological and

56 histological analysis revealed no malignancies or hyperplasia in any of the 44 endometrial samples. The diagnoses were as follows: 11 cases of atrophy or inactive endometrium, 16 cases of proliferation, 13 cases of secretion, three cases of desquamation and one case of polypoid endometrium.

5.2 Cost-minimization

Balloon endometrial ablation device was used in 15 of the 20 daycase procedures, and the remaining 5 were performed with the bipolar radiofrequency device, which was the method that was evaluated. The difference in time from the beginning of the procedure to the end between a balloon and bipolar procedure was a mean of 7 minutes, so that was deducted from the time spent on balloon procedures by all staff before making the calculations. Total costs for daycase procedure were 1,865 euros versus 1,065 euros for outpatient procedure, so the cost saved when the procedure was performed as an outpatient procedure under local anaesthetic instead of performing it as a daycase procedure in the operating theatre under general anaesthetic was 800 euros. The biggest saving was in staff costs. When endometrial ablation is performed in theatre, two operative nurses, one anaesthesia nurse and an anaesthesiologist are needed in addition to the gynaecologist. In an outpatient procedure two nurses and a gynaecologist are needed. The overhead costs are also bigger in a daycase procedure than in an outpatient procedure, 353 euros vs. 77 euros, respectively.

5.3 Effect of hyaluronic acid gel in prevention of adhesions

In a pilot study on the effect of hyaluronic acid gel in prevention of adhesions there were 4 women out of 32 who did not come to control hysteroscopy after endometrial ablation and one technically failed procedure. Of the 5 dropouts, 2 were in the study group and 3 were in the control group, so there were 14 women in the study group and 13 women in control group left for analysis. There was only one woman with a normal uterine cavity in the hysteroscopy, and she was in the study group. Tube-like cavity with or without adhesions was found in 6 women in the study group and in 5 women in the control group. Total obliteration was found in 7 women in the study group and in 8 women in the control group. Postoperative infection was reported in two women, one from the study group and another from the control group. Hyaluronic acid gel did not seem to prevent intrauterine adhesion formation after endometrial ablation.

57

58 6 Discussion

The first-line treatment of heavy menstrual bleeding is medical treatment with either tranexam acid, oral contraceptives or levonorgestrel-releasing intrauterine device. If the treatment fails, the next option is either hysterectomy or endometrial ablation. Sometimes women do not want to have intrauterine device for personal reasons; in our long-term study only 37% of women had tried treatment of HMB with LNG-IUS. Patients presenting with menorrhagia are typically 35 years or older, and many of them have gone through tubal ligation for birth control, and may thus be reluctant to use LNG-IUS to treat bleeding. Hysterectomy is a definitive solution, but it is a major operation with possible major complications. If a woman has a normal uterus which does not descend, the mode of hysterectomy is usually laparoscopic hysterectomy, which has a total complication rate of 15.4% and major complication rate of 4.3% (Brummer et al. 2011). The number of hysterectomies performed because of menstrual disorders in Finland declined until 2007, after which it has been the same until 2011, between 600 and 800 per year. It is assumed that 1/3 of hysterectomies performed for heavy menstrual bleeding are performed to women with normal uterus.

6.1 Long-term effects of endometrial ablation

In our retrospective study, the percentage of women who had a subsequent hysterectomy after endometrial ablation was comparable to previous studies (Amso et al. 2003, Amso 2006, Cooper & Gimpelson 2004). During the follow- up hysterectomy was performed to 16% of women, mostly for some other reason than solely menorrhagia. Some women had irregular , dysmenorrhea, indefinite pelvic pain or uterine fibroids preoperatively, but unfortunately the information concerning these pre-existing conditions was inadequate. Irregular periods are usually due to hormonal reasons, which can not be affected by endometrial ablation. Dysmenorrhea which is not limited to the menstrual cycle can be associated with adenomyosis, , fibroids or pelvic inflammatory disease, and responds to endometrial ablation more poorly than dysmenorrhea, which is restricted to pelvic cramping associated with the menstrual cycle. Previous studies have shown that young age, prior sterilization and history of dysmenorrhea are risk factors for treatment failure (El-Nashar et al. 2009, Bongers et al. 2002, Gemer et al. 2003, Longinotti et al. 2008, Mall et al. 2002). In our study these factors did not appear to be statistically significant, but

59 there was a similar trend. Submucotic myomas and endometrial polyps predict poor outcome of endometrial ablation, but also large intramural myomas may eventually lead to hysterectomy. As we found that some of our study patients had myomas preoperatively, it can be assumed that the number of hysterectomies could be reduced with better patient selection. The uterine cavity should be assessed preoperatively by some reliable means: ultrasound, sonohysterogram or by office hysteroscopy. Patient satisfaction is a very important endpoint in treating benign conditions like menorrhagia. In our study it was fairly high: 76% of the women were satisfied or fairly satisfied. This shows that improvement in quality of life is more important to women than amenorrhea.

6.2 Obliteration of the uterine cavity

Maybe the biggest problem concerning endometrial ablation is the tendency for uterine cavity obliteration postoperatively. In our observational study almost one in every four of the women had total obliteration of the cavity 6 years after EA, and outpatient endometrial sampling was impossible. All women had an endometrial sample successfully taken preoperatively. Endometrial sampling also failed with some women with regular menstruation. The explanation for this is that blood can come out through a channel that is narrower than is needed for endometrial sampling. The obliteration has no clinical relevance unless there is some endometrium trapped behind the obstruction. As residual endometrium almost always exists after endometrial ablation, there are some situations when endometrial sample should be retrieved, for example in the case of dysfunctional or postmenopausal bleeding or if a thick endometrium is seen on ultrasound scan of a postmenopausal woman. In those cases endometrial sampling may be possible under general anaesthesia, but there will be a risk of uterine perforation. It is also possible that the difficulty in getting diagnostic endometrial samples will lead to hysterectomy. However, despite being a common long-term consequence, the obliteration of the uterine cavity does not always cause problems. Only 5 of the 172 patients had hysterectomy because of pain, and dysmenorrhea was reported by 9.8% of the women, and we do not know how many of these conditions have developed after EA. Fortunately there is no evidence that diagnosis of endometrial cancer would be delayed because of this phenomenon (Krogh et al. 2009, Neuwirth et al. 2004, Valle & Baggish 1998).

60 There is some evidence that the degree of postoperative uterine adhesion formation is related to the treatment outcome (Friberg et al. 1998). As there is more amenorrhea after radiofrequency ablation than balloon ablation, it may be possible that there is also more total obliteration of uterine cavities after radiofrequency ablation. In our short-term study, in up to 61.5% of women there was no access to the uterine cavity 3 months after radiofrequency bipolar ablation. That is not of course comparable with long-term results 6 years after balloon ablation, but may show some tendency. Central hematometra and postablation tubal sterilization syndrome (PATSS) are possible consequences of trapped endometrial tissue. They are painful conditions which often need to be treated with hysterectomy. The diagnosis of PATSS can only be placed by serially sectioning the corneal area and proximal fallopian tubes of the removed uterus, and the pathologist should specifically be asked to do that. In our long-term study, diagnosis of PATSS was not placed in any of the 5 women who ended up with hysterectomy because of pain during the follow-up, which may be due to inability to recognize the condition. PATSS could be prevented with more efficient treatment of the corneal area. The fan-shaped device of the bipolar radiofrequency endometrial ablation method is designed to treat also the corneal area, so it is possible that there is less PATSS after radiofrequency ablation than balloon ablation, but there are no published data concerning this subject. Hyaluronic acid gel has been shown to prevent intrauterine adhesion formation after resection of submucotic myomas or uterine septas. However, in our randomized trial it did not have that effect after endometrial ablation. The sample size of our study was too small to find any statistically significant differences between groups, but even if there seemed to be a tendency towards less uterine cavity obliteration in the study group, the effect was not as expected according to previous studies. The reason is that the scarring with formation of fibrosis in the uterine cavity may continue even 4 to 5 months after endometrial ablation, but the hyaluronic acid gel stays in the uterine cavity only for approximately a week. Application of intrauterine device into the cavity after endometrial ablation for a few months could be a way to prevent uterine walls from collapsing together, but the efficacy and patient compliance remain to be studied. Therefore, this problem remains unsolved.

61 6.3 Economics

Novasure® is the most expensive of all EA devices, but it has the shortest treatment time and thus probably the best patient acceptability when performed under local anaesthetic. Patient acceptability of an outpatient EA has been shown to be good in many studies (Andersson & Mints 2007, Chapa 2008, Clark & Gupta 2004, Clark et al. 2011, Cooper 2005, Marsh et al. 2005, Marsh et al. 2007, Marsh et al. 2008, Penninx et al. 2009, Prasad & Powell 2008, Varma et al. 2010, Wallage et al. 2003), so we decided not to focus on that issue and did not collect information about patients’ experiences. In our experience, if a patient does not feel much inconvenience when an outpatient endometrial sample is taken, she will accept endometrial ablation under paracervical block and intravenous alfentanil well. Despite this, it could have been good to evaluate patient satisfaction and experience of pain also in our study. An interesting option for PCB could be a sublingual fentanyl buccal tablet, which is already used in some hospitals as an analgesic in outpatient hysteroscopies and colonoscopies. It could produce a rapid-onset and long enough-acting analgesia, offering even more cost and time savings, but there are as yet no trials concerning this. In our study we reduced the time consumed in thermal balloon ablations in the theatre to evaluate the costs of radiofrequency endometrial ablations in different settings. Because FREA is a mean of 7 minutes faster than TBA, it would be possible to perform more FREA-procedures than TBA-procedures per day, which could reduce the overhead costs. This means that the cost difference between settings might actually be a little smaller than in our calculations In our study we did not compare the costs of hysterectomy and endometrial ablation. In 2012 in Oulu University Hospital the price for laparoscopic hysterectomy, which in our hospital is the mode of hysterectomy when there is no prolapse, was 3,170 euros including 2 days at the ward. The price for an outpatient radiofrequency endometrial ablation (Novasure®) performed under local anaesthetic was 1,065 euros, so the cost saving is significant. Considering the savings to society related to the shorter sick leave of EA compared to that of hysterectomy, and the cost savings for patients, it can be stated that outpatient EA offers a cost-effective alternative to hysterectomy.

62 7 Conclusions

Endometrial ablation with second-generation devices has a good long-term effect in treatment of heavy menstrual bleeding. Intrauterine adhesion formation is common and does not seem to be avoided with hyaluronic acid gel. Adhesion formation causes significant difficulties in later evaluation of the endometrium, but according to this study or the literature, this does not seem to cause any severe problems. In our study the follow-up time is too short to evaluate the incidence or diagnostics of endometrial cancer, however. Performing the procedure as an outpatient procedure instead of a day case procedure in the operating theatre is significantly cost-saving. Endometrial ablation offers a good alternative for the treatment of heavy menstrual bleeding, and it seems to be especially suitable for women with a normal uterus who are approaching menopause.

63

64 References

Abbott J, Hawe J, Hunter D & Garry R (2003) A double-blind randomized trial comparing the Cavaterm and the NovaSure endometrial ablation systems for the treatment of dysfunctional uterine bleeding. Fertil Steril 80(1): 203–208. Aletebi FA, Vilos GA & Eskandar MA (1999) Thermal balloon endometrial ablation to treat menorrhagia in high-risk surgical candidates. J Am Assoc Gynecol Laparosc 6(4): 435–439. AlHilli MM, Hopkins MR & Famuyide AO (2011) Endometrial cancer after endometrial ablation: systematic review of medical literature. J Minim Invasive Gynecol 18(3): 393–400. Amin-Hanjani S & Good JM (1995) Pyometra after endometrial resection and ablation. Obstet Gynecol 85(5 Pt 2): 893–894. Amso NN (2006) Clinical and health service implications of second generation endometrial ablation devices. Curr Opin Obstet Gynecol 18(4): 457–463. Amso NN, Fernandez H, Vilos G, Fortin C, McFaul P, Schaffer M, Van der Heijden PF, Bongers MY, Sanders B & Blanc B (2003) Uterine endometrial thermal balloon therapy for the treatment of menorrhagia: long-term multicentre follow-up study. Hum Reprod 18(5): 1082–1087. Andersson S & Mints M (2007) Thermal balloon ablation for the treatment of menorrhagia in an outpatient setting. Acta Obstet Gynecol Scand 86(4): 480–483. Appiah-Sakyi K, Byrd L & Slade RJ (2004) Thermal balloon ablation-a rare case of fibroid necrosis. J Obstet Gynaecol 24(4): 467. Areia A, Branco M, Frutuoso C & de Oliveira CF (2006) Endometrial adenocarcinoma after endometrial ablation. A case report. Eur J Gynaecol Oncol 27(4): 432–433. Asherman JG (1950) Traumatic intra-uterine adhesions. J Obstet Gynaecol Br Em 57(6):892–6 Banu NS & Manyonda IT (2005) Alternative medical and surgical options to hysterectomy. Best Pract Res Clin Obstet Gynaecol 19(3): 431–449. Barrington JW, Arunkalaivanan AS & Abdel-Fattah M (2003) Comparison between the levonorgestrel intrauterine system (LNG-IUS) and thermal balloon ablation in the treatment of menorrhagia. Eur J Obstet Gynecol Reprod Biol 108(1): 72–74. Basak S & Saha A (2009) Adenomyosis: still largely under-diagnosed. J Obstet Gynaecol 29(6): 533–535. Baskett TF, Clough H & Scott TA (2005) NovaSure bipolar radiofrequency endometrial ablation: report of 200 cases. J Obstet Gynaecol Can 27(5): 473–476. Bhatia K, Doonan Y, Giannakou A & Bentick B (2008) A randomised controlled trial comparing GnRH antagonist cetrorelix with GnRH agonist leuprorelin for endometrial thinning prior to transcervical resection of endometrium. BJOG 115(10): 1214–1224.

65 Bhattacharya S, Middleton LJ, Tsourapas A, Lee AJ, Champaneria R, Daniels JP, Roberts T, Hilken NH, Barton P, Gray R, Khan KS, Chien P, O'Donovan P, Cooper KG, International Heavy Menstrual Bleeding Individual Patient Data Meta-analysis Collaborative Group, Abbott J, Barrington J, Bhattacharya S, Bongers MY, Brun JL, Busfield R, Clark TJ, Cooper J, Cooper KG, Corson SL, Dickersin K, Dwyer N, Gannon M, Hawe J, Hurskainen R, Meyer WR, O'Connor H, Pinion S, Sambrook AM, Tam WH, van Zon-Rabelink IA & Zupi E (2011) Hysterectomy, endometrial ablation and Mirena(R) for heavy menstrual bleeding: a systematic review of clinical effectiveness and cost-effectiveness analysis. Health Technol Assess 15(19): iii-xvi, 1–252. Bhattacharya S, Parkin DE, Reid TM, Abramovich DR, Mollison J & Kitchener HC (1995) A prospective randomised study of the effects of prophylactic antibiotics on the incidence of bacteraemia following hysteroscopic surgery. Eur J Obstet Gynecol Reprod Biol 63(1): 37–40. Bongers MY, Bourdrez P, Heintz AP, Brolmann HA & Mol BW (2005) Bipolar radio frequency endometrial ablation compared with balloon endometrial ablation in dysfunctional uterine bleeding: impact on patients' health-related quality of life. Fertil Steril 83(3): 724–734. Bongers MY, Bourdrez P, Mol BW, Heintz AP & Brolmann HA (2004) Randomised controlled trial of bipolar radio-frequency endometrial ablation and balloon endometrial ablation. BJOG 111(10): 1095–1102. Bongers MY, Mol BW & Brolmann HA (2002) Prognostic factors for the success of thermal balloon ablation in the treatment of menorrhagia. Obstet Gynecol 99(6): 1060–1066. Bourdrez P, Bongers MY & Mol BW (2004) Treatment of dysfunctional uterine bleeding: patient preferences for endometrial ablation, a levonorgestrel-releasing intrauterine device, or hysterectomy. Fertil Steril 82(1): 160–6, quiz 265. Bowling MR & Ramsey PS (2010) Spontaneous uterine rupture in pregnancy after endometrial ablation. Obstet Gynecol 115(2 Pt 2): 405–406. Brooks-Carter GN, Killackey MA & Neuwirth RS (2000) Adenocarcinoma of the endometrium after endometrial ablation. Obstet Gynecol 96(5 Pt 2): 836–837. Brown PM, Farquhar CM, Lethaby A, Sadler LC & Johnson NP (2006) Cost-effectiveness analysis of levonorgestrel intrauterine system and thermal balloon ablation for heavy menstrual bleeding. BJOG 113(7): 797–803. Brummer TH, Jalkanen J, Fraser J, Heikkinen AM, Kauko M, Makinen J, Seppala T, Sjoberg J, Tomas E & Harkki P (2011) FINHYST, a prospective study of 5279 hysterectomies: complications and their risk factors. Hum Reprod 26(7): 1741–1751. Busfield RA, Farquhar CM, Sowter MC, Lethaby A, Sprecher M, Yu Y, Sadler LC, Brown P & Johnson N (2006) A randomised trial comparing the levonorgestrel intrauterine system and thermal balloon ablation for heavy menstrual bleeding. BJOG 113(3): 257–263.

66 Carey ET, El-Nashar SA, Hopkins MR, Creedon DJ, Cliby WA & Famuyide AO (2011) Pathologic characteristics of hysterectomy specimens in women undergoing hysterectomy after global endometrial ablation. J Minim Invasive Gynecol 18(1): 96– 99. Carter JF (2005) Endometrial ablation: more choices, more options. The Female Patient 30(12):35–40. Chapa HO (2008) Utility of in-office endometrial ablation: a prospective cohort study of endometrial ablation under local anesthesia. J Reprod Med 53(11): 827–831. Clark TJ & Gupta JK (2004) Outpatient thermal balloon ablation of the endometrium. Fertil Steril 82(5): 1395–1401. Clark TJ, Samuel N, Malick S, Middleton LJ, Daniels J & Gupta JK (2011) Bipolar radiofrequency compared with thermal balloon endometrial ablation in the office: a randomized controlled trial. Obstet Gynecol 117(1): 109–118. Clarke A, Black N, Rowe P, Mott S & Howle K (1995) Indications for and outcome of total abdominal hysterectomy for benign disease: a prospective cohort study. Br J Obstet Gynaecol 102(8): 611–620. Clegg JP, Guest JF & Hurskainen R (2007) Cost-utility of levonorgestrel intrauterine system compared with hysterectomy and second generation endometrial ablative techniques in managing patients with menorrhagia in the UK. Curr Med Res Opin 23(7): 1637–1648. Colgan TJ, Shah R & Leyland N (1999) Post-hysteroscopic ablation reaction: a histopathologic study of the effects of electrosurgical ablation. Int J Gynecol Pathol 18(4): 325–331. Comino R & Torrejon R (2004) Hysterectomy after endometrial ablation-resection. J Am Assoc Gynecol Laparosc 11(4): 495–499. Cooper J, Gimpelson R, Laberge P, Galen D, Garza-Leal JG, Scott J, Leyland N, Martyn P & Liu J (2002) A randomized, multicenter trial of safety and efficacy of the NovaSure system in the treatment of menorrhagia. J Am Assoc Gynecol Laparosc 9(4): 418–428. Cooper J & Gimpelson RJ (2004) Summary of safety and effectiveness data from FDA: a valuable source of information on the performance of global endometrial ablation devices. J Reprod Med 49(4): 267–273. Cooper JM & Brady RM (2000) Late complications of operative hysteroscopy. Obstet Gynecol Clin North Am 27(2): 367–374. Cooper K, Lee A, Chien P, Raja E, Timmaraju V & Bhattacharya S (2011) Outcomes following hysterectomy or endometrial ablation for heavy menstrual bleeding: retrospective analysis of hospital episode statistics in Scotland. BJOG 118(10): 1171– 1179. Cooper KG (2005) Outpatient endometrial ablation. Best Pract Res Clin Obstet Gynaecol 19(5): 713–726. Daniels JP, Middleton LJ, Champaneria R, Khan KS, Cooper K, Mol BWJ, Bhattacharya S (2012) Second generation endometrial ablation techniques for heavy menstrual bleeding: network meta-analysis. BMJ 344:e2564.

67 Das S, Kirwan J, Drakeley AJ & Kingsland CR (2005) Pelvic abscess following microwave endometrial ablation. BJOG 112(1): 118–119. Daud S & Ewies AA (2008) Levonorgestrel-releasing intrauterine system: why do some women dislike it? Gynecol Endocrinol 24(12): 686–690. de Souza SS, Camargos AF, de Rezende CP, Pereira FA, Araujo CA & Silva Filho AL (2010) A randomized prospective trial comparing the levonorgestrel-releasing intrauterine system with thermal balloon ablation for the treatment of heavy menstrual bleeding. Contraception 81(3): 226–231. Deb S, Flora K, Atiomo W (2008) A survey of preferences and practices of endometrial ablation/resection for menorrhagia in the United Kingdom. Fertil Steril 90(5): 1812– 1817. DeCherney A & Polan ML (1983) Hysteroscopic management of intrauterine lesions and intractable uterine bleeding. Obstet Gynecol 61(3): 392–397. DeCherney AH, Diamond MP, Lavy G & Polan ML (1987) Endometrial ablation for intractable uterine bleeding: hysteroscopic resection. Obstet Gynecol 70(4): 668–670. Dickersin K, Munro MG, Clark M, Langenberg P, Scherer R, Frick K, Zhu Q, Hallock L, Nichols J, Yalcinkaya TM & Surgical Treatments Outcomes Project for Dysfunctional Uterine Bleeding (STOP-DUB) Research Group (2007) Hysterectomy compared with endometrial ablation for dysfunctional uterine bleeding: a randomized controlled trial. Obstet Gynecol 110(6): 1279–1289. Donnadieu AC, Deffieux X, Gervaise A, Faivre E, Frydman R & Fernandez H (2007) Essure sterilization associated with endometrial ablation. Int J Gynaecol Obstet 97(2): 139–142. Elmardi A, Furara S, Khan F & Hamza M (2009) NovaSure impedance controlled system for endometrial ablation: the experience of the first UK reference centre. J Obstet Gynaecol 29(5): 419–422. El-Nashar SA, Hopkins MR, Barnes SA, Pruthi RK, Gebhart JB, Cliby WA & Famuyide AO (2010) Health-related quality of life and patient satisfaction after global endometrial ablation for menorrhagia in women with bleeding disorders: a follow-up survey and systematic review. Am J Obstet Gynecol 202(4): 348.e1–348.e7. El-Nashar SA, Hopkins MR, Creedon DJ, Cliby WA & Famuyide AO (2009a) Efficacy of bipolar radiofrequency endometrial ablation vs thermal balloon ablation for management of menorrhagia: A population-based cohort. J Minim Invasive Gynecol 16(6): 692–699. El-Nashar SA, Hopkins MR, Creedon DJ, St Sauver JL, Weaver AL, McGree ME, Cliby WA & Famuyide AO (2009b) Prediction of treatment outcomes after global endometrial ablation. Obstet Gynecol 113(1): 97–106. El-Nashar SA, Hopkins MR, Feitoza SS, Pruthi RK, Barnes SA, Gebhart JB, Cliby WA & Famuyide AO (2007) Global endometrial ablation for menorrhagia in women with bleeding disorders. Obstet Gynecol 109(6): 1381–1387. El-Toukhy T, Chandakas S, Grigoriadis T, Hill N & Erian J (2004) Outcome of the first 220 cases of endometrial balloon ablation using Cavaterm plus. J Obstet Gynaecol 24(6): 680–683.

68 Ewies AA (2009) Levonorgestrel-releasing intrauterine system--the discontinuing story. Gynecol Endocrinol 25(10): 668–673. Fadare O, Qin L, Martel M & Tavassoli FA (2005) Pathology of the NovaSure (radio- frequency) impedance-controlled endometrial ablation system. Arch Pathol Lab Med 129(9): 1175–1178. Fakih M, Cherfan V & Abdallah E (2011) Success rate, quality of life, and descriptive analysis after generalized endometrial ablation in an obese population. Int J Gynaecol Obstet 113(2): 120–123. Friberg B, Joergensen C & Ahlgren M (1998) Endometrial thermal --degree of uterine fibrosis predicts treatment outcome. Gynecol Obstet Invest 45(1): 54–57. Fritsch H (1894) Ein Fall von volligem Schwaund der Gebormutterhohle nach Auskratzung. Zentralbl Gynaekol 18:1337–1342. Fulop T, Rakoczi I & Barna I (2007) NovaSure impedance controlled endometrial ablation: long-term follow-up results. J Minim Invasive Gynecol 14(1): 85–90. Gallinat A (2004) NovaSure impedance controlled system for endometrial ablation: three- year follow-up on 107 patients. Am J Obstet Gynecol 191(5): 1585–1589. Gallinat A (2007) An impedance-controlled system for endometrial ablation: five-year follow-up of 107 patients. J Reprod Med 52(6): 467–472. Gallinat A & Nugent W (2002) NovaSure impedance-controlled system for endometrial ablation. J Am Assoc Gynecol Laparosc 9(3): 283–289. Gangadharan A, Revel A & Shushan A (2010) Endometrial thermal balloon ablation in women with previous cesarean delivery: pilot study. J Minim Invasive Gynecol 17(3): 358–360. Garside R, Stein K, Wyatt K & Round A (2005) Microwave and thermal balloon ablation for heavy menstrual bleeding: a systematic review. BJOG 112(1): 12–23. Garside R, Stein K, Wyatt K, Round A & Pitt M (2004) A cost-utility analysis of microwave and thermal balloon endometrial ablation techniques for the treatment of heavy menstrual bleeding. BJOG 111(10): 1103–1114. Garza-Leal J, Hernandez I, Castillo L, Jessop ML, Chill N & Coad JE (2011) Successful vapor-based endometrial ablation: in vivo peri-hysterectomy study. J Minim Invasive Gynecol 18(4): 445–448. Gemer O, Kapustian V, Kroll D, Tur-Kaspa I & Segal S (2003) Perioperative factors for predicting successful hysteroscopic endometrial ablation. J Reprod Med 48(9): 677– 680. Gervaise A, de Tayrac R & Fernandez H (2005) Contraceptive information after endometrial ablation. Fertil Steril 84(6): 1746–1747. Giarenis I, Shenoy J & Morris E (2008) Cervical ectopic pregnancy after endometrial ablation: a case report. Arch Gynecol Obstet 277(6): 567–569. Glasser MH, Heinlein PK & Hung YY (2009) Office endometrial ablation with local anesthesia using the HydroThermAblator system: Comparison of outcomes in patients with submucous myomas with those with normal cavities in 246 cases performed over 5(1/2) years. J Minim Invasive Gynecol 16(6): 700–707.

69 Glasser MH & Zimmerman JD (2003) The HydroThermAblator system for management of menorrhagia in women with submucous myomas: 12- to 20-month follow-up. J Am Assoc Gynecol Laparosc 10(4): 521–527. Goldberg J, McCrosson S & Kaulback KR (2005) Delayed leiomyoma degeneration after microwave endometrial ablation. Obstet Gynecol 106(5 Pt 2): 1176–1178. Goldrath MH, Fuller TA & Segal S (1981) Laser photovaporization of endometrium for the treatment of menorrhagia. Am J Obstet Gynecol 140(1): 14–19. Goni AZ, Lacruz RL, Paricio JJ & Hernandez Rivas FJ (2009) The levonorgestrel intrauterine system as an alternative to hysterectomy for the treatment of idiopathic menorrhagia. Gynecol Endocrinol 25(9): 581–586. Gurtcheff SE & Sharp HT (2003) Complications associated with global endometrial ablation: the utility of the MAUDE database. Obstet Gynecol 102(6): 1278–1282. Hardwick JC & Owen P (2004) Adherence to published guidelines for the management of menorrhagia in women undergoing second generation endometrial ablation. J Obstet Gynaecol 24(3): 279–280. Hill DJ (1994) Hematometra-A Complication of Endometrial Ablation/Resection. J Am Assoc Gynecol Laparosc 1(4, Part 2): S14. Hillis SD, Marchbanks PA, Tylor LR & Peterson HB (1998) Higher hysterectomy risk for sterilized than nonsterilized women: findings from the U.S. Collaborative Review of Sterilization. The U.S. Collaborative Review of Sterilization Working Group. Obstet Gynecol 91(2): 241–246. Hoffman MK & Sciscione AC (2004) Placenta accreta and intrauterine fetal death in a woman with prior endometrial ablation: a case report. J Reprod Med 49(5): 384–386. Huchon C & Fritel X (2008) Epidemiology of . J Gynecol Obstet Biol Reprod (Paris) 37 Suppl 8: S307–16. Hurskainen R (2006) Managing drug-resistant essential menorrhagia without hysterectomy. Best Pract Res Clin Obstet Gynaecol 20(5): 681–694. Hurskainen R, Teperi J, Rissanen P, Aalto AM, Grenman S, Kivela A, Kujansuu E, Vuorma S, Yliskoski M & Paavonen J (2004) Clinical outcomes and costs with the levonorgestrel-releasing intrauterine system or hysterectomy for treatment of menorrhagia: randomized trial 5-year follow-up. JAMA 291(12): 1456–1463. Iliodromiti S & Murage A (2011) Multiple bowel perforations requiring extensive bowel resection and hysterectomy after microwave endometrial ablation. J Minim Invasive Gynecol 18(1): 118–120. Istre O, Forman A & Bourne TH (1996a) The relationship between preoperative endometrial thickness, the anteroposterior diameter of the uterus and clinical outcome following transcervical resection of the endometrium. Ultrasound Obstet Gynecol 8(6): 412–416. Istre O, Holm-Nielsen P, Bourne T & Forman A (1996b) Hormone replacement therapy after transcervical resection of the endometrium. Obstet Gynecol 88(5): 767–770.

70 Jack SA, Cooper KG, Seymour J, Graham W, Fitzmaurice A & Perez J (2005) A randomised controlled trial of microwave endometrial ablation without endometrial preparation in the outpatient setting: patient acceptability, treatment outcome and costs. BJOG 112(8): 1109–1116. Jamieson R, Hammond I & Maouris P (2002) Small bowel perforation associated with microwave endometrial ablation. Aust N Z J Obstet Gynaecol 42(4): 407–408. Jarrell A & Olsen ME (2003) Patient satisfaction with thermal balloon endometrial ablation. A retrospective review. J Reprod Med 48(8): 635–636. Jarvela I, Tekay A, Santala M & Jouppila P (2002) Ultrasonographic features following thermal balloon endometrial ablation therapy. Gynecol Obstet Invest 54(1): 11–16. Kaunitz AM, Meredith S, Inki P, Kubba A & Sanchez-Ramos L (2009) Levonorgestrel- releasing intrauterine system and endometrial ablation in heavy menstrual bleeding: a systematic review and meta-analysis. Obstet Gynecol 113(5): 1104–1116. Khan Z, El-Nashar SA, Hopkins MR & Famuyide AO (2011) Efficacy and safety of global endometrial ablation after cesarean delivery: a cohort study. Am J Obstet Gynecol . Kilonzo MM, Sambrook AM, Cook JA, Campbell MK & Cooper KG (2010) A cost-utility analysis of microwave endometrial ablation versus thermal balloon endometrial ablation. Value Health 13(5): 528–534. Kleijn JH, Engels R, Bourdrez P, Mol BW & Bongers MY (2008) Five-year follow up of a randomised controlled trial comparing NovaSure and ThermaChoice endometrial ablation. BJOG 115(2): 193–198. Kopeika J, Edmonds SE, Mehra G & Hefni MA (2011) Does hydrothermal ablation avoid hysterectomy? Long-term follow-up. Am J Obstet Gynecol 204(3): 207.e1–207.e8. Krogh RA, Lauszus FF, Guttorm E & Rasmussen K (2009) Surgery and cancer after endometrial resection. Long-term follow-up on menstrual bleeding and hormone treatment by questionnaire and registry. Arch Gynecol Obstet 280(6): 911–916. Kuzel D, Bartosova L, Rezabek K, Toth D, Cindr J & Mara M (2010) Successful pregnancy after thermal balloon endometrial ablation followed by in vitro fertilization and embryo transfer. Fertil Steril 93(3): 1006.e11–1006.e13. Laberge PY, Sabbah R, Fortin C & Gallinat A (2003) Assessment and comparison of intraoperative and postoperative pain associated with NovaSure and ThermaChoice endometrial ablation systems. J Am Assoc Gynecol Laparosc 10(2): 223–232. Le Marrec A, Lavoue V, Morcel K, Duval H, Bauville E, Foucher F & Leveque J (2010) Occurrence of endometrial cancer six years after treatment with thermal balloon ablation (Thermachoice): first case report. Eur J Obstet Gynecol Reprod Biol 150(2): 219–220. Lee JK, Gupta J, McCarthy K & Gornall RJ (2005) Endometrial adenocarcinoma following ThermaChoice balloon endometrial ablation. J Obstet Gynaecol 25(8): 825– 826. Lete I, Obispo C, Izaguirre F, Orte T, Rivero B, Cornellana MJ, Bermejo I & Spanish Society of Gynaecology Obstetrics (SEGO) (2008) The levonorgestrel intrauterine system (Mirena) for treatment of idiopathic menorrhagia. Assessment of quality of life and satisfaction. Eur J Contracept Reprod Health Care 13(3): 231–237.

71 Lethaby A, Hickey M, Garry R & Penninx J (2009) Endometrial resection / ablation techniques for heavy menstrual bleeding. Cochrane Database Syst Rev (4): CD001501. Lethaby A, Shepperd S, Cooke I & Farquhar C (2000) Endometrial resection and ablation versus hysterectomy for heavy menstrual bleeding. Cochrane Database Syst Rev (2): CD000329. Leung PL, Tam WH & Yuen PM (2003) Hysteroscopic appearance of the endometrial cavity following thermal balloon endometrial ablation. Fertil Steril 79(5): 1226–1228. Leung PL & Yuen PM (2006) Postablation-tubal sterilization syndrome following thermal balloon endometrial ablation. Acta Obstet Gynecol Scand 85(4): 504–505. Levgur M (2007) Therapeutic options for adenomyosis: a review. Arch Gynecol Obstet 276(1): 1–15. Lo JS & Pickersgill A (2006) Pregnancy after endometrial ablation: English literature review and case report. J Minim Invasive Gynecol 13(2): 88–91. Loffer FD (2006) Endometrial ablation in patients with myomas. Curr Opin Obstet Gynecol 18(4): 391–393. Lok IH, Leung PL, Ng PS & Yuen PM (2003) Life-table analysis of the success of thermal balloon endometrial ablation in the treatment of menorrhagia. Fertil Steril 80(5): 1255–1259. Longinotti MK, Jacobson GF, Hung YY & Learman LA (2008) Probability of hysterectomy after endometrial ablation. Obstet Gynecol 112(6): 1214–1220. Luo X, Lim CE, Li L & Wong WS (2010) Hysteroscopic appearance of endometrial cavity after microwave endometrial ablation. J Minim Invasive Gynecol 17(1): 30–36. Luoto R, Kaaja R, Raitanen J (2003) Naisten terveys – kuukautiset, raskauskomplikaatiot, ehkäisy, hormonikorvaushoidot ja lapsettomuus FINRISKI 2002–tutkimuksessa. taulukkoraportti. Kansanterveyslaitoksen julkaisuja B18 Magos AL, Baumann R & Turnbull AC (1989) Transcervical resection of endometrium in women with menorrhagia. BMJ 298(6682): 1209–1212. Mall A, Shirk G & Van Voorhis BJ (2002) Previous tubal ligation is a risk factor for hysterectomy after rollerball endometrial ablation. Obstet Gynecol 100(4): 659–664. Marret H, Fauconnier A, Chabbert-Buffet N, Cravello L, Golfier F, Gondry J, Agostini A, Bazot M, Brailly-Tabard S, Brun JL, De Raucourt E, Gervaise A, Gompel A, Graesslin O, Huchon C, Lucot JP, Plu-Bureau G, Roman H, Fernandez H & CNGOF College National des Gynecologues et Obstetriciens Francais (2010) Clinical practice guidelines on menorrhagia: management of abnormal uterine bleeding before menopause. Eur J Obstet Gynecol Reprod Biol 152(2): 133–137. Marsh F, Bekker H & Duffy S (2008) A survey of women's views of Thermachoice endometrial ablation in the outpatient versus day case setting. BJOG 115(1): 31–37. Marsh F, Thewlis J & Duffy S (2005) Thermachoice endometrial ablation in the outpatient setting, without local anesthesia or intravenous sedation: a prospective cohort study. Fertil Steril 83(3): 715–720. Marsh F, Thewlis J & Duffy S (2007) Randomized controlled trial comparing Thermachoice III* in the outpatient versus daycase setting. Fertil Steril 87(3): 642– 650.

72 McCausland AM & McCausland VM (1999) Partial rollerball endometrial ablation: a modification of total ablation to treat menorrhagia without causing complications from intrauterine adhesions. Am J Obstet Gynecol 180(6 Pt 1): 1512–1521. McCausland AM & McCausland VM (2002) Frequency of symptomatic cornual hematometra and postablation tubal sterilization syndrome after total rollerball endometrial ablation: a 10-year follow-up. Am J Obstet Gynecol 186(6): 1274–80; discussion 1280–3. McCausland AM & McCausland VM (2007) Long-term complications of endometrial ablation: cause, diagnosis, treatment, and prevention. J Minim Invasive Gynecol 14(4): 399–406. McPherson K, Herbert A, Judge A, Clarke A, Bridgman S, Maresh M & Overton C (2005) Psychosexual health 5 years after hysterectomy: population-based comparison with endometrial ablation for dysfunctional uterine bleeding. Health Expect 8(3): 234–243. McPherson K, Metcalfe MA, Herbert A, Maresh M, Casbard A, Hargreaves J, Bridgman S & Clarke A (2004) Severe complications of hysterectomy: the VALUE study. BJOG 111(7): 688–694. Mettler L (2002) Long-term results in the treatment of menorrhagia and hypermenorrhea with a thermal balloon endometrial ablation technique. LSLS 6(4): 305–309. Middleton LJ, Champaneria R, Daniels JP, Bhattacharya S, Cooper KG, Hilken NH, O'Donovan P, Gannon M, Gray R, Khan KS, International Heavy Menstrual Bleeding Individual Patient Data Meta-analysis Collaborative Group, Abbott J, Barrington J, Bhattacharya S, Bongers MY, Brun JL, Busfield R, Sowter M, Clark TJ, Cooper J, Cooper KG, Corson SL, Dickersin K, Dwyer N, Gannon M, Hawe J, Hurskainen R, Meyer WR, O'Connor H, Pinion S, Sambrook AM, Tam WH, van Zon-Rabelink IA & Zupi E (2010) Hysterectomy, endometrial destruction, and levonorgestrel releasing intrauterine system (Mirena) for heavy menstrual bleeding: systematic review and meta-analysis of data from individual patients. BMJ 341: c3929. Mircea CN, Goojha C & Thiel JA (2011) Concomitant NovaSure endometrial ablation and Essure tubal sterilization: a review of 100 cases. J Obstet Gynaecol Can 33(4): 361– 366. Mishra K, Manucha V, Sengupta R, Singh K & Aggarwal N (2003) Cytomorphology and histomorphology after thermal endometrial ablation. Early changes. Acta Cytol 47(6): 1001–1007. Moorman PG, Myers ER, Schildkraut JM, Iversen ES, Wang F & Warren N (2011) Effect of hysterectomy with ovarian preservation on ovarian function. Obstet Gynecol 118(6): 1271–1279. Mukul LV & Linn JG (2005) Pregnancy complicated by uterine synechiae after endometrial ablation. Obstet Gynecol 105(5 Pt 2): 1179–1182. Munro M (2004) Endometrial ablation with a thermal balloon: the first 10 years. J Am Assoc Gynecol Laparosc 11(1): 8–22. Munro MG (2006) Endometrial ablation: where have we been? Where are we going? Clin Obstet Gynecol 49(4): 736–766.

73 Nagele F, Rubinger T & Magos A (1998) Why do women choose endometrial ablation rather than hysterectomy? Fertil Steril 69(6): 1063–1066. National Institute for Health and Clinical Excellence (2007) Heavy menstrual bleeding; quick reference guide. NICE clinical guideline 44. Neuwirth RS, Loffer FD, Trenhaile T & Levin B (2004) The incidence of endometrial cancer after endometrial ablation in a low-risk population. J Am Assoc Gynecol Laparosc 11(4): 492–494. Olson S, Wallage S, Deans HE, Wallis F & Parkin DE (2002) Magnetic resonance imaging appearances of the uterus following microwave endometrial ablation. Clin Radiol 57(10): 926–929. Onoglu A, Taskin O, Inal M, Sadik S, Simsek M, Akar M, Kursun S, Mendilcioglu I, Postaci H & Ispahi C (2007) Comparison of the long-term histopathologic and morphologic changes after endometrial rollerball ablation and resection: a prospective randomized trial. J Minim Invasive Gynecol 14(1): 39–42. Overton C, Hargreaves J & Maresh M (1997) A national survey of the complications of endometrial destruction for menstrual disorders: the MISTLETOE study. Minimally Invasive Surgical Techniques--Laser, EndoThermal or Endorescetion. Br J Obstet Gynaecol 104(12): 1351–1359. Penninx JP, Mol BW & Bongers MY (2009) Endometrial ablation with paracervical block. J Reprod Med 54(10): 617–620. Penninx JP, Mol BW, Engels R, van Rumste MM, Kleijn C, Koks CA, Kruitwagen RF & Bongers MY (2010) Bipolar radiofrequency endometrial ablation compared with hydrothermablation for dysfunctional uterine bleeding: a randomized controlled trial. Obstet Gynecol 116(4): 819–826. Perez-Medina T, Bajo-Arenas J, Sanfrutos L, Haya J, Iniesta S & Vargas J (2003) Endometrial intraepithelial neoplasia diagnosed at endometrial resection. J Am Assoc Gynecol Laparosc 10(4): 542–544. Practice Committee of American Society for Reproductive Medicine (2008) Indications and options for endometrial ablation. Fertil Steril 90(5 Suppl): S236–40. Prasad P & Powell MC (2008) Prospective observational study of Thermablate Endometrial Ablation System as an outpatient procedure. J Minim Invasive Gynecol 15(4): 476–479. Preutthipan S & Herabutya Y (2010) Hysteroscopic rollerball endometrial ablation as an alternative treatment for adenomyosis with menorrhagia and/or dysmenorrhea. J Obstet Gynaecol Res 36(5): 1031–1036. Reid PC (2007) Endometrial ablation in England--coming of age? An examination of hospital episode statistics 1989/1990 to 2004/2005. Eur J Obstet Gynecol Reprod Biol 135(2): 191–194. Reid PC, Thurrell W, Smith JH, Kennedy A & Sharp F (1992) Nd:YAG laser endometrial ablation: histological aspects of uterine healing. Int J Gynecol Pathol 11(3): 174–179.

74 Roberts TE, Tsourapas A, Middleton LJ, Champaneria R, Daniels JP, Cooper KG, Bhattacharya S & Barton PM (2011) Hysterectomy, endometrial ablation, and levonorgestrel releasing intrauterine system (Mirena) for treatment of heavy menstrual bleeding: cost effectiveness analysis. BMJ 342: d2202. Robinson R, China S, Bunkheila A & Powell M (2008) Mirena intrauterine system in the treatment of menstrual disorders: a survey of UK patients' experience, acceptability and satisfaction. J Obstet Gynaecol 28(7): 728–731. Robson S & Devine B (2007) Two cases of leiomyoma necrosis after thermal balloon endometrial ablation. J Minim Invasive Gynecol 14(2): 250–252. Rogerson L & Duffy S (2002) A European survey of the complications of a uterine thermal balloon ablation system in 5800 women. Gynaecological Endoscopy (11): 171–176. Rooney KE & Cholhan HJ (2010) Vesico-uterine fistula after endometrial ablation in a woman with prior cesarean deliveries. Obstet Gynecol 115(2 Pt 2): 450–451. Roth TM & Rivlin ME (2004) Tuboovarian abscess: a postoperative complication of endometrial ablation. Obstet Gynecol 104(5 Pt 2): 1198–1199. Roy KH & Mattox JH (2002) Advances in endometrial ablation. Obstet Gynecol Surv 57(12): 789–802. Sabbah R & Desaulniers G (2006) Use of the NovaSure Impedance Controlled Endometrial Ablation System in patients with intracavitary disease: 12-month follow- up results of a prospective, single-arm clinical study. J Minim Invasive Gynecol 13(5): 467–471. Sagiv R, Ben-Shem E, Condrea A, Glezerman M & Golan A (2005) Endometrial carcinoma after endometrial resection for dysfunctional uterine bleeding. Obstet Gynecol 106(5 Pt 2): 1174–1176. Salmeen K & Morgan D (2009) Sepsis after bipolar radiofrequency endometrial ablation. Obstet Gynecol 114(2 Pt 2): 445–448. Sambrook AM, Cooper KG, Campbell MK & Cook JA (2009) Clinical outcomes from a randomised comparison of Microwave Endometrial Ablation with Thermal Balloon endometrial ablation for the treatment of heavy menstrual bleeding. BJOG 116(8): 1038–1045. Sambrook AM, Jack SA & Cooper KG (2010) Outpatient microwave endometrial ablation: 5-year follow-up of a randomised controlled trial without endometrial preparation versus standard day surgery with endometrial preparation. BJOG 117(4): 493–496. Schlumbrecht M, Balgobin S & Word L (2007) Pyometra after thermal endometrial ablation. Obstet Gynecol 110(2 Pt 2): 538–540. Sesti F, Ruggeri V, Pietropolli A, Piancatelli R & Piccione E (2011) Thermal balloon ablation versus laparoscopic supracervical hysterectomy for the surgical treatment of heavy menstrual bleeding: a randomized study. J Obstet Gynaecol Res 37(11): 1650– 1657. Seymour J, Wallage S, Graham W, Parkin D & Cooper K (2003) The cost of microwave endometrial ablation under different anaesthetic and clinical settings. BJOG 110(10): 922–926.

75 Shaamash AH & Sayed EH (2004) Prediction of successful menorrhagia treatment after thermal balloon endometrial ablation. J Obstet Gynaecol Res 30(3): 210–216. Shamonki MI, Ziegler WF, Badger GJ & Sites CK (2000) Prediction of endometrial ablation success according to perioperative findings. Am J Obstet Gynecol 182(5): 1005–1007. Shapley M, Jordan K & Croft PR (2004) An epidemiological survey of symptoms of menstrual loss in the community. Br J Gen Pract 54(502): 359–363. Sharp HT (2012) Endometrial ablation: postoperative complications. Am J Obstet Gynecol . Shaw RW, Symonds IM, Tamizian O, Chaplain J & Mukhopadhyay S (2007) Randomised comparative trial of thermal balloon ablation and levonorgestrel intrauterine system in patients with idiopathic menorrhagia. Aust N Z J Obstet Gynaecol 47(4): 335–340. Sinervo K & Martyn P (2000) Endometrial stromal sarcoma diagnosed after hysteroscopic endometrial resection. J Am Assoc Gynecol Laparosc 7(2): 257–259. Sowter MC, Lethaby A & Singla AA (2002) Pre-operative endometrial thinning agents before endometrial destruction for heavy menstrual bleeding. Cochrane Database Syst Rev (3)(3): CD001124. Sowter MC (2003) New surgical treatments for menorrhagia. Lancet 361: 1456–58 Suomen Gynekologiyhdistys ry:n asettama työryhmä (2005) Runsaiden kuukautisvuotojen tutkimus ja hoito. Duodecim 121(14): 1584–92. Taskin O, Onoglu A, Inal M, Turan E, Sadik S, Vardar E, Postaci H & Wheeler JM (2002) Long-term histopathologic and morphologic changes after thermal endometrial ablation. J Am Assoc Gynecol Laparosc 9(2): 186–190. Thinkhamrop J, Laopaiboon M & Lumbiganon P (2007) Prophylactic antibiotics for transcervical intrauterine procedures. Cochrane Database Syst Rev (3)(3): CD005637. Townsend DE, McCausland V, McCausland A, Fields G & Kauffman K (1993) Post- ablation-tubal sterilization syndrome. Obstet Gynecol 82(3): 422–424. Tresserra F, Grases P, Ubeda A, Pascual MA, Grases PJ & Labastida R (1999) Morphological changes in hysterectomies after endometrial ablation. Hum Reprod 14(6): 1473–1477. Turnbull LW, Jumaa A, Bowsley SJ, Dhawan S, Horsman A & Killick SR (1997) Magnetic resonance imaging of the uterus after endometrial resection. Br J Obstet Gynaecol 104(8): 934–938. US Department of Health and Human Services, US Food and Drug Administration (2011) MAUDE-manufacturer and user facility device experience. Valle RF & Baggish MS (1998) Endometrial carcinoma after endometrial ablation: high- risk factors predicting its occurrence. Am J Obstet Gynecol 179(3 Pt 1): 569–572. Valle RF, Valdez J, Wright TC & Kenney M (2006) Concomitant Essure tubal sterilization and Thermachoice endometrial ablation: feasibility and safety. Fertil Steril 86(1): 152–158. van der Vaart CH, van der Bom JG, de Leeuw JR, Roovers JP & Heintz AP (2002) The contribution of hysterectomy to the occurrence of urge and stress urinary incontinence symptoms. BJOG 109(2): 149–154.

76 Vancaillie TG (1989) Electrocoagulation of the endometrium with the ball-end resectoscope. Obstet Gynecol 74(3 Pt 1): 425–427. Varma R, Soneja H, Samuel N, Sangha E, Clark TJ & Gupta JK (2010) Outpatient Thermachoice endometrial balloon ablation: long-term, prognostic and quality-of-life measures. Gynecol Obstet Invest 70(3): 145–148. Vaughan D & Byrne P (2012) An evaluation of the simultaneous use of the levonorgestrel- releasing intrauterine device (LNG-IUS, Mirena(R)) combined with endometrial ablation in the management of menorrhagia. J Obstet Gynaecol 32(4): 372–374. Wallage S, Cooper KG, Graham WJ & Parkin DE (2003) A randomised trial comparing local versus general anaesthesia for microwave endometrial ablation. BJOG 110(9): 799–807. You JH, Sahota DS & MoYuen P (2006) A cost-utility analysis of hysterectomy, endometrial resection and ablation and medical therapy for menorrhagia. Hum Reprod 21(7): 1878–1883.

77

78 Original publications

I Ahonkallio S, Martikainen H & Santala M (2008) Endometrial thermal balloon has a beneficial effect on menorrhagia. Acta Obstet Gynecol Scand 87: 107–110 II Ahonkallio S, Liakka A, Martikainen H, Santala M (2009) Feasibility of endometrial assessment after thermal ablation. Eur J Obstet Gynecol Reprod Biol 147: 69–71 III Ahonkallio S, Santala M, Valtonen H, Martikainen H (2012) Cost-minimisation analysis of endometrial thermal ablation in a daycase or outpatient setting under different anaesthesia regimens. Eur J Obstet Gynecol Reprod Biol 162: 102–104 IV Ahonkallio S, Martikainen H, Santala M (2012) Auto-crosslinked hyaluronic acid gel in prevention of intrauterine adhesions after bipolar radiofrequency endometrial ablation. A randomised, controlled and double blind pilot study. A manuscript. Reprinted with permission from John Wiley and Sons (I) and Elsevier (II,III).

Original publications are not included in the electronic version of the dissertation.

79

80 ACTA UNIVERSITATIS OULUENSIS SERIES D MEDICA

1189. Hookana, Eeva (2012) Characteristics of victims of non-ischemic sudden cardiac death 1190. Murugan, Subramanian (2012) Control of nephrogenesis by Wnt4 signaling : mechanisms of gene regulation and targeting of specific lineage cells by tissue engineering tools 1191. Nikkilä, Jenni (2013) PALB2 and RAP80 genes in hereditary breast cancer predisposition 1192. Määttä, Mikko (2012) Assessment of osteoporosis and fracture risk : axial transmission ultrasound and lifestyle-related risk factors 1193. Kaakinen, Marika (2012) Genetic and life course determinants of cardiovascular risk factors : structural equation modelling of complex relations 1194. Nevalainen, Mika (2012) Gene product targeting into and membrane trafficking from the endoplasmic/sarcoplasmic reticulum in skeletal myofibers 1195. Meriläinen, Sanna (2013) Experimental study of acute pancreatitis in a porcine model, especially tight junction structure and portal vein cytokines 1196. Koivikko, Minna (2013) Cardiac autonomic regulation and repolarisation during hypoglycaemia in type 1 diabetes 1197. Aro, Ellinoora (2013) Prolyl 4-hydroxylases, key enzymes regulating hypoxia response and collagen synthesis : the roles of specific isoenzymes in the control of erythropoiesis and skeletogenesis 1198. Koivumäki, Janne (2013) Biomechanical modeling of proximal femur : development of finite element models to simulate fractures 1199. Paavola, Liisa (2013) Salla disease – rare but diverse : a clinical follow-up study of a Finnish patient sample 1200. Vesterinen, Soili (2013) Osastonhoitajien johtamistyylit osana johtamiskulttuuria 1201. Huusko, Tuija (2013) Genetic and molecular background of ascending aortic aneurysms 1202. Pisto, Pauliina (2013) Fat accumulation in liver and muscle : Association with adipokines and risk of cardiovascular events 1203. Yannopoulos, Fredrik (2013) Remote ischemic preconditioning as a means to protect the brain against hypothermic circulatory arrest : An experimental study on piglets 1204. Arvonen, Miika (2013) Intestinal immune activation in Juvenile idiopathic arthritis Book orders: Granum: Virtual book store http://granum.uta.fi/granum/ D 1205 OULU 2013 D 1205

UNIVERSITY OF OULU P.O.B. 7500 FI-90014 UNIVERSITY OF OULU FINLAND ACTA UNIVERSITATIS OULUENSIS ACTA UNIVERSITATIS OULUENSIS ACTA

SERIES EDITORS DMEDICA Sari Ahonkallio

ASCIENTIAE RERUM NATURALIUM Sari Ahonkallio Senior Assistant Jorma Arhippainen ENDOMETRIAL THERMAL BHUMANIORA ABLATION University Lecturer Santeri Palviainen CTECHNICA A CHOICE FOR TREATMENT OF HEAVY Docent Hannu Heusala MENSTRUAL BLEEDING DMEDICA Professor Olli Vuolteenaho ESCIENTIAE RERUM SOCIALIUM University Lecturer Hannu Heikkinen FSCRIPTA ACADEMICA Director Sinikka Eskelinen GOECONOMICA Professor Jari Juga

EDITOR IN CHIEF Professor Olli Vuolteenaho PUBLICATIONS EDITOR

Publications Editor Kirsti Nurkkala UNIVERSITY OF OULU GRADUATE SCHOOL; UNIVERSITY OF OULU, FACULTY OF MEDICINE, ISBN 978-952-62-0142-9 (Paperback) INSTITUTE OF CLINICAL MEDICINE, ISBN 978-952-62-0143-6 (PDF) DEPARTMENT OF OBSTETRICS AND GYNECOLOGY; ISSN 0355-3221 (Print) OULU UNIVERSITY HOSPITAL ISSN 1796-2234 (Online)