Institute Clinic of Gynaecology, Obstetrics and Neonatology, Faculty of Medicine, University of Barcelona, Hospital Clinic-Institut d′Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain
Dr C Castelo-Branco, Institut Clínic de Ginecologia, Obstetrícia i Neonatología, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain. Email firstname.lastname@example.org
Please cite this paper as: Povedano B, Arjona J, Velasco E, Monserrat J, Lorente J, Castelo-Branco C. Complications of hysteroscopic Essure® sterilisation: report on 4306 procedures performed in a single centre. BJOG 2012;119:795–799.
Objective To analyse the short-, medium- and long-term complications in women undergoing hysteroscopic tubal sterilisation with the Essure® device.
Design Retrospective 7-year study.
Setting Office hysteroscopic unit in a teaching hospital.
Sample A total of 4306 women whoe underwent the Essure® sterilisation procedure from 2003 to 2010.
Methods Data on the success of the procedure and complications arising from outpatient hysteroscopic sterilisation using the Essure® system were collected from consecutive women undergoing the procedure over a 7-year period.
Main outcome measures Placement rate, successful bilateral tubal occlusion, perioperative adverse events, early postoperative (during the first 3 months of follow-up) and late complications (after the initial 3 months of follow-up).
Results A total of 4108 (96.8%) women completed the standard 3-month follow-up protocol. Only 534 (13%) women had undergone the procedure within the previous year. There were 115 (out of 4306; 2.7%) recorded complications, none of which resulted in the need for hospitalisation or discharge later than 2 hours after the procedure. Vasovagal syncope was the most frequently encountered adverse event, occurring in 85 (2.0%) of 4306 cases. In 19 cases, one device was expelled, with most expulsions (14 out of 19) being detected before or during the 3-month follow-up.
Conclusions Outpatient hysteroscopic sterilisation using the Essure® system is safe, with a low rate of complications.
Hysteroscopic sterilisation using Essure® (Conceptus, Mountain View, CA, USA) has been used increasingly throughout the world, especially within the European Union, after the technology was introduced in 2001. Compared with classic laparoscopic sterilisation techniques (bipolar coagulation, Hulka clips, Filshie clips and silicone rings), hysteroscopic sterilisation can be performed with less surgical time in an outpatient clinic setting, without the need for surgical incisions or general anaesthesia, and with superior patient tolerance and patient satisfaction.1–4 However, the ability to complete the procedure is generally lower than traditional laparoscopic approaches. The transcervical route of accessing the fallopian tubes requires proficiency in outpatient hysteroscopy as well as familiarity with the Essure® technique. Failure to satisfactorily insert the devices into the fallopian tubes varies from 88 to 98%,1,4,5 although successful placement appears to be improving as experience amongst gynaecologists increases in conjunction with continued improvements in Essure® technology.6
To date, complications associated with the Essure® procedure are thought to be uncommon, and are minor when present.7 On the other hand, complications with laparoscopy, although rare, can be serious, including organ or vascular injury and problems arising from general anaesthesia.8,9 Possible adverse events with the Essure® procedure are related either to the general hysteroscopic approach or the specific Essure® placement technique. Complications such as fluid overload and electrolyte disturbances are unusual, as the distending media for Essure® sterilisation is saline solution, and the operating time on average decreases to around 8 minutes, with experience.10,11 Uterine and cervical trauma is rare, as the procedure is completed with small-diameter continuous-flow hysteroscopes. In a phase-II study, 7% of women reported an adverse event, of which vasovagal syncope was the most prevalent, followed by device expulsion, perforation and unsatisfactory placement.5 Pain during and after the procedure, bleeding and dyspareunia have also been described.5 Persistent pain has been reported, but is unusual and, if present, raises the possibility of tubal perforation by the device, which may affect around 1–2% of procedures.12,13 Migration to the cavity can also occur without tubal perforation.14 Confirmatory radiological testing with abdominal X-ray, transvaginal ultrasound (TVU) or hysterosalpingogram (HSG), according to local protocols, is mandatory to establish satisfactory device placement and/or tubal occlusion. Early expulsion of a device from the tube and uterine cavity is usually diagnosed at this time, but if this takes place after the 3-month follow-up test it may remain undetected, leaving the patient at risk of an unwanted pregnancy.15
There is a need for more data regarding the feasibility and safety of hysteroscopic sterilisation. We therefore undertook a study to analyse the practicality and short-, medium- and long-term complications of hysteroscopic tubal sterilisation using the Essure® permanent birth control system.
All women attending the Andalusia Health Service in Cordoba requesting permanent birth control, and who underwent hysteroscopic sterilisation with the Essure® system at the ‘Reina Sofía’ University Hospital between March 2003 and June 2010 were identified. A total of 4306 women were included in this study carried out from March 2003 to June 2010. The inclusion criteria for hysteroscopic sterilisation were women aged 18–45 years desiring permanent birth control and able to comprehend and sign written consent, attend the 3permanent birth control, and who underwent hysteroscopic sterilisation with the Essmonth follow-up and use contraception until then. Women were excluded if they had abnormal uterine bleeding, active pelvic inflammatory disease or uncertainty about their desire to end their fertility. All procedures were carried out or supervised by four gynaecologists (B.P., E.V., J.M. and J.L.) experienced in the Essure® technique. Of the 4306 procedures, 1683 were performed by resident gynaecologists and 293 were performed by external training gynaecologists.
The procedures were scheduled to be performed in the follicular phase of the menstrual cycle, unless women were taking oral contraceptives, in which case they were advised to take their pills continuously for at least a month prior to the intervention. All women received premedication with oral ibuprofen and 10 mg of diazepam 1 hour before the procedure. Of the sample, 472 women received paracervical anaesthesia (mepivacaine cloridrate 3%, 7.2 ml) because they found routine gynaecological examination painful, had a tight cervical os or were anxious.1 A vaginoscopic approach to hysteroscopy was adopted routinely, and the Essure® microinserts were placed through the 5F working channel of a 5-mm, continuous-flow, 30° hysteroscope (Karl Storz Endoscopia, Barcelona, Spain), using a saline distension medium. Placement was performed according to the manufacturer’s instructions. Placement was considered to be optimal when between one and eight coils remained visible in the uterine cavity. After insertion, the women consulted and completed an oral questionnaire defining their pain experience as none, mild (less than menses), moderate (similar to menses) or severe (more than menses). All women who underwent Essure® placement were discharged home on the same day of the procedure.
Women were advised to use an alternative contraceptive method until a simple pelvic X-ray examination was performed at least 3 months after the insertion. Placement was considered satisfactory when both devices appeared to be within the tubal lumen, had symmetrical appearance, ends that looked opposite one another and when the distance between the intrauterine ends was not >4 cm. If the position of the devices were not clear in the pelvic X-ray a TVU was performed with recourse to HSG if the ultrasound was equivocal (i.e. if the transversal section of the uterus did not show both devices at the uterotubal junction). In addition, HSG was performed when the placement was not satisfactory (more than eight or less than one coils remaining visible by hysteroscopy, insertion only in one tube or highly difficult procedure). On confirmation of adequate placement at the 3-month follow-up visit, women were advised to rely on the Essure® system for contraception.
Adverse events were recorded at the time of the procedure, at the 3-month follow-up and at each gynaecological check-up. The telephone number of the unit was given to all the women, who were advised to report any complication. All general practitioners and general gynaecologists in the healthcare reference area were advised to report and refer the subject in the case of complications being detected in Essure® users.
The mean age of the sample was 36 years, ranging from 19 to 49 years, with 1306 women aged 40–49 years, 2687 aged 30–39 years, 315 aged 20–29 years and three women who were <20 years old. Of the 4306 women undergoing hysteroscopic Essure® sterilisation, 4242 (98.5%) were successfully completed, of which 4075 were successfully sited at one attempt (96%), without the need for a second-stage procedure at a later date. A total of 4108 (96.8%) women complied with the 3-month radiological follow-up. The majority of women had follow-up beyond 1 year, with only 534 cases having had the procedure within the previous year (Figure 1). The rate of successful bilateral obstruction (including unilateral occlusion in women with only one tube) was 99.7% (4095/4108). One or both tubes remained patent in 13 women (0.3%), representing a 3.1:1000 failure rate (Figure 2).
A total of seven women (0.16%), three before and four after the 3-month follow-up, became pregnant. The pregnancy rate after assessment for satisfactory placement was 0.09%. In one subject who became pregnant 32 months after the procedure, the 3-month control pelvic X-ray and vaginal ultrasound scan suggested that the devices were well placed, and the HSG performed after delivery showed an apparent bilateral occlusion. However, laparoscopy demonstrated a unilateral tubal perforation.
The most frequent perioperative complication was vasovagal syncope, occurring in 85 women (1.9%). Of 19 device expulsions (0.4%), 14 occurred during the early postoperative period; in 13 women another device was placed, and in two other women the expulsion was incomplete, and in these cases the coils trailing into the cavity were cut. Subsequent HSG demonstrated correct bilateral tubal obstruction. In three cases the devices were erroneously placed in the myometrium (0.06%). In two other cases, asymptomatic migrations into the abdominal cavity (0.04%) were detected, as were two cases of pelvic inflammatory disease (0.04%). Both women with abdominal migration of one device underwent another placement, retaining the migrated devices in the abdominal cavity. Longer term complications included one woman who suffered persistent abdominal pain (0.02%) and one women with tubal perforation (0.02%). Two allergies to nickel were noted. The first case was a woman with a history of atopy, who presented shortly after the procedure with papular urticaria and erythema, and in whom the symptoms disappeared after the subsequent removal of both devices. The other case was a woman referred with a history of persistent genital pruritus 1 year after the placement who decided to remove the devices.
In 3568 women (82.8%), the perioperative pain was considered non-existent or mild, which was managed with oral non-steroidal anti-inflammatory drugs (NSAIDS). Moderate pain was reported by 513 (11.9%) women, and severe pain was reported by 155 women (3.5%). In 70 cases (1.5%) pain was not recorded at all. One subject experienced persistent abdominal pain that did not disappear with NSAIDS. As the pain remained after the removal of the devices, a prophylactic appendicectomy and cholecystectomy was performed, but the pain continued to persist. This woman is currently receiving treatment with antidepressant drugs.
This study has demonstrated Essure® hysteroscopic sterilisation to be a successful and safe procedure. No adverse events identified were associated with serious long-term medical problems. Indeed 75% of complications were minor, self-limiting episodes of vasovagal syncope, a phenomenon associated with any form of uterine instrumentation, and not specific to the Essure® procedure. Interestingly, a higher incidence of vasovagal syncope was observed in interventions performed by gynaecologists with a shorter learning curve.1 Techniques aimed at reducing the incidence of vasovagal syncope include performing the procedure with smaller diameter instruments (avoiding the use of the outflow channel) and reducing the intrauterine pressure provided by the perfusion pump to 90 mmHg on the appearance of vasovagal symptoms, although there is no evidence to support these manoeuvres at present.
No serious sequelae arose from the abnormal placement of the device. Among the 19 device expulsions recorded, another device was placed in 13 of the women, and 3 months later the HSG performed to assess the location of the device demonstrated secondary bilateral tube obstructions. In two other cases, the expulsion was partially towards the endometrial cavity, and the coils seen in the cavity were cut, showing a correct bilateral obstruction on HSG. In the two cases of cavity migration a new device was placed without the removal of the migrated devices. Both patients remain asymptomatic, without pain, infection symptoms or gastrointestinal symptoms. Nickel allergies affect 17% of women,16 representing 721 cases of expected nickel allergies in 4242 placements. Nonetheless, in this series only two women presented with an allergy to nickel. This low incidence could be explained by the very low proportion of nickel in the devices, and thus allergy symptoms only develop in a very small number of patients who are extremely sensitive to this element. Pelvic inflammatory disease is uncommon after hysteroscopy. Only two cases of pelvic inflammatory disease were observed in this series, both of which resolved after treatment with endovenous antibiotics. Our results are in agreement with those described by others. In a prospective study including 1952 surgical hysteroscopies, only 18 endometritis and 12 urinary tract infections, but no cases of pelvic inflammatory disease, were observed.17 Another recent study reported two cases (79%) of pelvic inflammatory disease after 253 diagnostic hysteroscopies.18 Therefore, prophylactic antibiotic treatment prior to the Essure® procedure is not recommended.
Our series supports the contention that Essure® hysteroscopic sterilisation is acceptable, with <5% of women considering the pain experienced during the procedure as severe, and is effective at preventing pregnancy, with only seven pregnancies reported in over 4000 procedures. Of these seven procedures that resulted in pregnancy, three women did not comply with the follow-up protocol and became pregnant prior to confirmatory radiological testing. Of the four true method failures (i.e. women who became pregnant after apparently successful procedures), one case is of note as it occurred over 3 years after the procedure. An HSG at 3 months had shown an apparent bilateral occlusion. The woman decided to continue with the pregnancy, and after delivery laparoscopy showed a right tubal perforation.19 This highlights the need for long-term data regarding the effectiveness of Essure® sterilisation in preventing pregnancy so as to better aid the counselling and informed consenting of our patients.
Our series is the largest to date evaluating the feasibility, safety and effectiveness of Essure® hysteroscopic sterilisation. We identified all women undergoing the procedure. However, the incidence of adverse events post-operatively may be an underestimate as we relied upon self-reporting. However, we did emphasise the importance of reporting any complications to our unit via telephone, and complications were enquired about at each consultation. Moreover, healthcare professionals in primary care and peripheral centres were encouraged to report any adverse events presumed to have arisen from the procedure. We relied upon pelvic X-ray examination to confirm the adequacy of sterilisation, although the accuracy of pelvic X-ray has been questioned compared with TVU or HSG.20 Whereas it is likely that some complications may have remained undetected, we believe that it is unlikely that any major adverse events, such as pelvic inflammatory disease or pregnancy, occurred without our knowledge.
In conclusion, our results, the largest series by far to date, support existing data demonstrating that outpatient clinic tubal hysteroscopic sterilisation with Essure® is feasible, successful and safe.
Disclosure of interests
The authors have received expenses from Conceptus (the manufacturer of Essure®) to attend scientific meetings and congresses.
Contribution to authorship
BP was involved in the conception and design of the study, performed procedures, managed the inclusion and follow-up of participants, collected and analysed data, and approved the final version of the article. JEA was involved in the conception and design of the study, performed procedures, managed the inclusion and follow-up of participants, collected and analyses data, and approved the final version of the article. EV performed procedures, managed the inclusion and follow-up of participants, collected data and approved the final version of the article. JAM performed procedures, managed the inclusion and follow-up of participants, collected data and approved the final version of the article. JL performed procedures, collected data, reviewed the article and approved the final version. CC-B was involved in the conception and design of the study, performed statistical analyses, performed the final analysis of the data, drafted and revised the article for important intellectual content, and approved the final version.
Details of ethics approval
The study was performed in a regular practice, and all the women demonstrated comprehension after the explanation of the procedure and signed a written consent that emphasised the irreversibility of the procedure, and included the possibility of failure to achieve sterilisation at first attempt. IRB Reina Sofia University Hospital. January 2003.