Female sterilisation is the most common method of contraception in the world,1 and laparoscopic sterilisation has been the preferred method for over 30 years. The procedure is simple, safe and effective, especially when mechanical devices such as the Filshie clip are used as opposed to electrocoagulation.2 Over the last decade, however, the supremacy of laparoscopic access to the fallopian tube has been challenged by the hysteroscopic route. The notion of non-incisional, transcervical access to the oviduct is not new, and indeed was first attempted in the early part of the 20th century, with more concerted efforts being made from the 1970s onwards. However, it was only in this century that a technology was developed that was safe, feasible, acceptable, reproducible and effective: the Essure permanent birth control system (Conceptus, Mountain View, CA, USA).3 This convenient, hysteroscopic procedure is well described, being performed as an outpatient procedure without anaesthesia in under 10 minutes, and with discharge from hospital or community-based clinics within an hour.3–5 Compare this truly ambulatory procedure with the rigmarole of current methods of laparoscopic sterilisation. The gynaecologist is in control of their service, bypassing the hurdles presented by waiting list bureaucracy, preoperative assessment, ward staff, theatre staff, porters and anaesthetists. Economic evaluations have consistently demonstrated the cost-effectiveness of the Essure system over the laparoscopic route, and women, given the choice between modalities, prefer it (T. J. Clark, pers. comm).
So, given the advantages of hysteroscopic sterilisation, why in 2012 is it not established in day-to-day gynaecological practice? Well, in contrast to laparoscopic sterilisation, hysteroscopic procedures require around 3 months to occlude the narrow, intramyometrial lumen of the fallopian tube. Additional contraceptive precautions are necessary until radiological testing confirms the adequacy of sterilisation: satisfactory device placement is checked by abdominal X-ray or transvaginal ultrasound scan in the 85% of procedures considered to be straightforward, and the remainder undergo more invasive hysterosalpingogram (HSG) to confirm tubal occlusion,5 although in the USA HSG is mandated. Over 500 000 Essure kits have been sold worldwide, but the uptake of the procedure varies markedly within and outside of national boundaries, and the reasons for this eclectic practice appears to be complex, relating to local cultures, expertise, infrastructure and health service funding. In the UK, around 500 hysteroscopic procedures were undertaken in 2010, compared with approximately 8500 laparoscopic sterilisations: a ratio of 17:1.6 Spain has a similar uptake of female sterilisation per head of population to the UK,1 but hysteroscopic sterilisation is more widely practiced, with one centre alone undertaking 600 procedures each year.7 Of course, the heterogeneous practice may also reflect a paucity of good-quality, large data sets supporting the use of this health technology. It is therefore timely that this issue of BJOG reports the largest single-centre study to date of outpatient hysteroscopic sterilisation.7 This Spanish study included over 4000 women and demonstrated the technique to be safe, feasible and effective in the short and medium term, with a median follow up of 3–4 years post-procedure.
What messages can we take from the paper by Castelo-Branco et al.?7 Firstly the procedure is practical and achievable in an outpatient setting. The standard measure of the successful completion of the office procedure is satisfactory bilateral placement of the Essure devices (‘micro inserts’) at the initial attempt. Smaller published series report bilateral placement rates of between 86 and 99%.5 The main impediment to placement is problematic visualisation of the tubal ostia, which can be more of a problem in the secretory phase of the menstrual cycle, as endometrial congestion and fragmentation is more likely then.4 Adverse anatomical factors, including the lateralisation of tubal ostia and proximal tubal stenosis or spasm, are impediments, as is to a lesser extent patient tolerance.4 The current series reports the highest bilateral placement rate of 96% at first attempt, increasing to 99% if one includes second-stage procedures. Given the size of the current series, the confidence limits around the estimate of successful placement are narrow, and if one takes into account the lower 95% confidence estimate (97%), it is still higher than most other reported series.
The reasons for this higher success rate are unclear. The population does not appear to be highly selected, and is generalisable to most settings. The protocol employed does not appear overly novel. The authors timed their procedure where possible in the proliferative phase, which has been shown to improve the likelihood of successful placement.7 However, most procedures in the secretory or menstrual phase are successful, and so it is unlikely that the scheduling of the procedure is the only factor explaining the success of this series. A vaginoscopic approach (i.e. without the use of a vaginal speculum or cervical instrumentation) is routinely employed elsewhere, and local infiltration of local anaesthesia into the cervix is standard practice if cervical dilatation is required. The use of conscious sedation is not, however, widespread. This group routinely used 10 mg of oral diazepam, but patient tolerance for procedure failure is less important than adverse intrauterine or tubal anatomy,5 which would not be affected by the use of sedation. One could speculate that in difficult cases, rather than abandoning procedures, tubal cannulation was achieved through perseverance or suboptimal placement accepted. If this were the case one would expect higher pain scores, reports of suboptimal placement, complications (e.g. perforation) or unplanned pregnancies. Data from the series do not support this contention. It may therefore reflect expertise within the centre, and thus generalisability may be limited. However, although four surgeons led the procedures, 46% were performed by gynaecologists in training. This is no mean feat, as will be appreciated by those of us who have tried to supervise trainees in outpatient hysteroscopic procedures! Moreover, an earlier publication by the same group of 857 procedures reported similar success rates,8 and so it does not appear to be a learning curve phenomenon. Thus, the successful placement rates presented here appear to be the benchmark that the rest of us should aim to achieve.
Laparoscopic sterilisation requires a general anaesthetic and entry (usually blind) into the abdominal cavity. The procedure most commonly employs the use of clips applied to the fallopian tube, and has proven its safety and effectiveness over the years.2 However, serious complications can arise from inadvertent damage to bowel, bladder and major blood vessels. As body mass indices increase and caesarean section becomes more prevalent, one would expect such complications to become more likely. The present series of over 4000 hysteroscopic sterilisations demonstrates the safety advantages of avoiding entry into the peritoneal cavity, with no major complications observed.7 It is probable that major complications will be reported upon the more widespread adoption of hysteroscopic sterilization, with the subsequent increase in the availability of longer term follow-up data.9 However, the absence of major injury in the biggest series to date is reassuring. The authors reported side effects in 115/4306 (3%) procedures, including 21 device expulsions, two device migrations, one tubal perforation, one case of pelvic inflammatory disease, one case of persistent pelvic pain and two presumed nickel allergies. The vasovagal syncope rate was only 2%, but this seems low when one compares with other series and personal experience (10% at the Birmingham Women’s Hospital; T. J. Clark, pers. comm.). This low incidence is likely to result from a lack of standardisation of definition. In any case, such reactions are usually minor and self-limiting, without the need for additional interventions, apart from basic supportive measures.
Currently the Royal College of Obstetricians and Gynaecologists (RCOG) recommend that we inform women that laparoscopic sterilisation has a failure rate of 1 in 200.10 In the current large series, the effectiveness of the procedure after successful device placement was 99.1% at a median follow-up time of 4 years (ranging from 3 months to 7 years). The pivotal trial data from the original female cohort has recorded no pregnancies to date,4 and the estimated cumulative 5-year effectiveness rate is 99.7%.11 Based upon the number of reported cases of unintended pregnancy worldwide (mandatory in the USA), divided by the number of Essure kits sold worldwide, corresponds to a failure rate of 0.15% (748/497 305; M. Vleugels, pers. comm.). Although this figure is likely to be an underestimate, the efficacy rate in general gynaecological practice, at least in the short and medium term, is consistent with the published series, and compares favourably with the leading laparoscopic technique in the UK.2,5
What about patient acceptability? The authors of the current series collected perioperative pain scores from all women undergoing the procedure: only 3.5% of women reported perioperative pain as being severe.7 These pain scores may have been subject to observer bias, as they were not collected independently from the treating clinician, but they are in line with other published data.3 The authors did not collect qualitative data addressing patient experience, which would have enhanced the article. However, with acceptable pain scores, short procedure times and a lack of perioperative complications, it seems reasonable to assume that women found the procedure acceptable, as has been previously reported.4
The recognition that hysteroscopic sterilisation is safe, convenient, acceptable and probably preferred by women over traditional approaches, has led to the development of alternative hysteroscopic technologies. The Adiana® system (Hologic, Inc., Bedford, MA, USA) has recently been introduced, and is the only competing hysteroscopic approach currently available on the market. In contrast to Essure sterilisation, Adiana placement requires the use of a non-electrolyte fluid distension media, and involves creating a small ‘lesion’ within the proximal fallopian tube using a 60-second burst of bipolar energy, after which a tiny micro-insert made of silicone matrix is then placed into the tube, and tissue grows into it over the course 3 months.12 An HSG is then required at 3 months. The feasibility of Adiana appears comparable with the Essure system. The reported pregnancy prevention rate is 98.4% at 3 years,12 which also compares favourably with traditional laparoscopic techniques.2 However, data regarding the effectiveness of the Adiana technique in general gynaecological practice is scarce compared with the more established Essure system. A major challenge facing current and future hysteroscopic approaches to permanent birth control relates to their global application. Costs are currently prohibitive for use of these technologies in low-resource settings, although they have been implemented sporadically, funded through charitable, global women’s healthcare organisations.
So, should we consign laparoscopic sterilisation to history? At the present time the answer has to be no, given that experience and practice of laparoscopic procedures is far more widespread, as are all long-term safety and cumulative effectiveness data. That said, the available data pertaining to safety, patient preference, acceptability, effectiveness and cost-effectiveness of the available hysteroscopic technologies appear to compare favourably with traditional laparoscopic sterilisation. The time for a large, direct randomised comparison has come, and such a study is supported by a recent survey of members of the British Society of Gynaecological Endoscopy (T. J. Clark, pers. comm.). The choice of the primary effectiveness outcome will be problematic, because unintended pregnancy is rare for both procedures, but successful procedure completion, patient satisfaction or health-related quality of life seem good surrogate outcome denominators, against which costs can be applied. Only then may we be able to counsel our patients in an informed, unbiased way, and allocate resources for this common gynaecological procedure in a rational manner.