Essure® for management of hydrosalpinx prior to in vitro fertilisation—a systematic review and pooled analysis

Authors


Abstract

Background

Hydrosalpinges in infertile women reduce the success of in vitro fertilisation (IVF) by 50%. Surgical management of hydrosalpinges before IVF improves outcome but these procedures are often contraindicated in women with dense pelvic adhesions. Tubal occlusion achieved by Essure® via hysteroscopy provides an alternative.

Objectives

To conduct a systematic review on the efficacy and safety of Essure® in the management of hydrosalpinx before IVF.

Search strategy

We searched MEDLINE (January 1950 to July 2013), EMBASE (January 1980 to July 2013) and Web of Science (1899 to July 2013). We also searched reference lists of relevant articles and proceedings of relevant international conferences (2000–2013).

Selection criteria

All types of studies where women with suspected infertility and presence of hydrosalpinx had hysteroscopic tubal occlusion with Essure® before IVF.

Data collection and analysis

Two authors independently selected studies and extracted data. Where necessary, study authors were contacted for further data.

Main results

In all, 115 women in 11 studies received Essure®, mainly in the outpatient setting where local anaesthesia by paracervical block and/or intravenous sedation was used. Successful placement of Essure® was achieved in 96.5% (95% confidence interval [95% CI] 91.1–98.9%) of women and tubal occlusion in 98.1% (95% CI 93.1–99.9%). Subsequent IVF resulted in 38.6% pregnancy rate (95% CI 30.9–46.8%), 27.9% live birth rate (95% CI 21.1–35.8%) and 28.6% combined ongoing pregnancy and live birth rate (95% CI 21.7–36.6%) per embryo transfer.

Author's conclusions

Essure® appears to be an effective option for management of hydrosalpinx in women before IVF although evidence from a randomised controlled clinical trial is lacking.

Introduction

Hydrosalpinges are found in 10–30% of couples presenting with infertility from tubal factors.[1-5] In addition to its role in infertility, hydrosalpinges have an adverse effect on the success of in vitro fertilisation (IVF). Live birth rates of infertile women with hydrosalpinges undergoing IVF are reduced by 50% in comparison to controls, i.e. infertile women without hydrosalpinges.[6, 7] Management of hydrosalpinges, usually by laparoscopic procedures (salpingectomy/proximal tubal occlusion), before IVF leads to improved outcomes.[8, 9] These procedures are invasive and carry the risk of organ or vessel injury especially in the presence of dense pelvic adhesions, which are seen in those with previous multiple laparotomies, severe endometriosis, Crohn's disease or ulcerative colitis. For such patient groups expected to have dense adhesions, laparoscopy can be challenging and high risk.

Alternative treatment options include aspiration of hydrosalpingeal fluid at the time of IVF procedure or achieving tubal occlusion of the hydrosalpinx by devices inserted hysteroscopically, which would avoid several of the above risks. Two such devices; Essure® (Conceptus Inc., San Carlos, CA, USA) and Adiana® (Hologic, Bedford, MA, USA) are known to achieve hysteroscopic tubal occlusion. Both have been widely used for sterilisation, but only Essure® has been used until recently in the context of hydrosalpinx. The first report of use of Adiana® for a similar indication has just been made. This review therefore focuses on the use of Essure®.[10]

Essure® is a microinsert that is 4 cm in length and 2 mm in diameter in its expanded form. Polyethylene terephthalate fibres run along through the inner coil and these fibres induce a tissue reaction resulting in tubal occlusion. The nickel titanium outer coil serves as an anchor within the uterotubal junction.[11] Essure® has been extensively used for hysteroscopic sterilisation; by the end of 2008 nearly 250 000 women were using Essure®.[11, 12] The role of the Essure® microinsert in hydrosalpinx is likely to be by a similar mechanism of tubal occlusion, which prevents hydrosalpingeal fluid from entering the endometrial cavity. In 2005, Rosenfield et al.[13] first reported a successful pregnancy (and subsequently live birth) in an obese woman (body mass index >50 kg/m2) with a hydrosalpinx and extensive pelvic adhesions following use of Essure® before IVF.

The main objective of the systematic review is to analyse the published and grey literature on the use of Essure® in the management of hydrosalpinx before IVF.

Methods

Criteria for considering studies for this review

Types of studies

All types of studies including case reports, case series, controlled clinical trials and randomised controlled trials.

Types of participants

All women with suspected infertility and presence of unilateral or bilateral hydrosalpinx.

Types of interventions

For case reports and case series, the intervention was Essure®. For controlled clinical trials and randomised controlled trials, the intervention was Essure® versus no intervention or standard practice or any other intervention.

Types of outcome measures

Outcomes were assessed in relation to two areas.

Procedure-related

  1. Successful placement of Essure® in each woman—assessed by direct visualisation. Any woman with bilateral hydrosalpinx was considered to have successful placement of Essure® if it was successfully placed in both fallopian tubes.
  2. Successful placement of Essure® in each fallopian tube—assessed by direct visualisation.
  3. Successful tubal occlusion in each woman—assessed by hysterosalpingogram unless specified otherwise. Any woman with bilateral hydrosalpinx was considered to have successful tubal occlusion if both fallopian tubes were occluded.
  4. Successful tubal occlusion of each fallopian tube—assessed by hysterosalpingogram unless specified otherwise.
  5. Complications of procedure.

IVF-related

  1. Pregnancy rate per embryo transfer excluding biochemical pregnancy where the only evidence of pregnancy was a short-lived rise of human chorionic gonadotrophin levels with no accompanying clinical or radiological evidence.
  2. Miscarriage rate per pregnancy.
  3. Live birth rate per embryo transfer.
  4. Preterm live birth rate per embryo transfer.
  5. Ongoing pregnancy and live birth rate per embryo transfer.

Search methods for identification of studies

Electronic searches

The following databases were searched: MEDLINE (January 1950 to July 2013), EMBASE (January 1980 to July 2013) and Web of Science (1899 to July 2013). The search strategy combined ‘essure’ or ‘microinsert’ with ‘hydrosalpinx’ or ‘hydrosalpinges’. There were no language restrictions.

Searching other sources

The following were also searched:

  1. Abstracts of the following conference proceedings were hand-searched: American Society for Reproductive Medicine 2000–2012, European Society of Human Reproduction and Embryology 2000–2013, and British Fertility Society 2000–2013.
  2. Reference lists of included studies were searched.
  3. Ongoing research trials were identified from the WHO International clinical Trials Registry Platform (http://apps.who.int/trialsearch/), metaRegister of Controlled Trials (www.controlled-trials.com/mrct/), EU Clinical Trials Register (www.clinicaltrialsregister.eu/) and the National Research Register (www.nihr.ac.uk/Pages/NRRArchiveSearch.aspx). If any relevant studies were found, the investigators were contacted to obtain more information.
  4. The UK sales representative of Essure® was contacted.

Data collection and analysis

The results from different databases were merged and duplicate reports of the same study were removed. The results of the search were independently screened by PA and RSA. They were not blinded to the journal, the authors or the institution. The first stage of screening of the search results involved assessing titles and abstracts to determine whether each article met the inclusion criteria. If the title and abstract did not provide enough information to make a decision then the full text of the article was obtained and the inclusion criteria were applied. Where appropriate, the authors of the individual studies were contacted to clarify study eligibility.

Specifically designed forms for data extraction were used. Data were independently extracted by PA and RSA. Where appropriate, the authors of the individual studies were contacted to obtain additional data. The data from the studies are presented in a tabular form. The data from each study were pooled to give a summary statistic for each of the outcomes. The confidence intervals (CI) for the pooled estimates were calculated using the Wald method.[14]

Assessment of risk of bias in included studies

PA and RSA independently assessed the methodological quality of selected studies. For randomised controlled clinical trials, the specific tool recommended by The Cochrane Collaboration was used[15] whereas for nonrandomised studies, including cohort studies, the Newcastle–Ottawa Scale was used.[16]

Results

Results of the search

The search strategy identified 48 studies, including 40 studies from electronic search of databases and an additional eight studies from proceedings of scientific meetings. Twenty-three were duplicates and were excluded. The remaining 25 studies were screened and 14 of these were excluded,[10, 17-29] the reasons for which are shown in Figure 1.

Figure 1.

PRISMA flow diagram for study selection. American Society for Reproductive Medicine (ASRM), British Fertility Society (BFS), European Society of Human Reproduction and Embryology (ESHRE).

Included studies

Eleven studies (eight published manuscripts and three abstracts from scientific meetings) were included in the review.[12, 30-39] These included three single case reports and eight case series. One of the studies had a control where the standard practice was laparoscopic salpingectomy for hydrosalpinx (control arm) unless surgery was contraindicated because of previous surgery and adhesions where Essure® was inserted (study arm).[39] The baseline features of these studies are listed in Table 1 and risk of bias assessment using the Newcastle–Ottawa Scale (as none of the studies had a randomised control) is shown in Table 2. Limited information was extracted from one included study. Contact with the author could not be made because correspondence details were not available.[32]

Table 1. Baseline features of included studies
Study IDLocationType of studyNumber of patients with hydrosalpinxAge (years)ComorbidityAnaesthesia used
TotalU/LB/L
  1. B/L, bilateral; BMI, body mass index; PID, pelvic inflammatory disease; U/L, unilateral.

  2. a

    Data missing.

  3. b

    Only data of patient on study arm i.e. Essure® included.

Rosenfield et al.[13]USACase report11031Obese (BMI 50) with pelvic adhesionsParacervical block (lidocaine) and light sedation (propofol)
Hitkari et al.[30]CanadaCase series52332–41All had pelvic adhesions (endometriosis, Crohn's disease, previous surgeries)General anaesthesia in all patients
Galen[31]USACase series72532–44None a
Gil and Arias[32]VenezuelaCase series26 a a 32–41 a Sedation used in six patients
Omurtag et al.[33]USACase report11037Pelvic adhesions a
Nichols and West[34]USACase series61534–41Two had previous abdominopelvic surgeriesFour had local anaesthesia and two had conscious sedation
Galen et al.[35]USACase series2081224–44NoneNine paracervical block (lidocaine), 11 intravenous sedation (midazolam, fentanyl, propofol)
Mijatovic et al.[36]NetherlandsCase series2013728–3811 had endometriosis, seven had PID, two had Crohn's disease)Five had paracervical block
Thébault et al.[37]FranceCase series1331024–39Four had endometriosis, two had pelvic adhesionsFive had sedation, one had spinal anaesthesia, seven had general anaesthesia
Inocêncio et al.[38]PortugalCase report11029 a a
Matorras et al.[39]SpainCase series with controlb156931–39All had pelvic adhesions (endometriosis, PID, tuberculosis)All had sedation with diazepam while one patient had paracervical block
Total  115385124–44  
Table 2. Risk of bias assessment using Newcastle–Ottawa assessment scale
 SelectionComparabilityOutcomeTotal stars
123456789
  1. (1) indicates the exposed cohort study representative of the population; (2) the nonexposed cohort drawn from the same population; (3) the exposure ascertainments are from secure record or structured interview; (4) the outcome was not present at start of study; (5) cohorts are comparable for underlying diagnosis; (6) cohorts are comparable for presence of adhesions; (7) assessment of outcome is from secure record; (8) follow up long enough for outcome to occur; and (9) complete follow up.

Rosenfield et al.[13]   6
Hitkari et al.[30]   6
Galen[31]   6
Gil and Arias[32]      3
Omurtag et al.[33]   6
Nichols and West[34]   6
Galen et al.[35]   6
Mijatovic et al.[36]   6
Thébault et al.[37]   6
Inocêncio et al.[38]   6
Matorras et al.[39] 8

A total of 115 women (median seven women per study, range 1–26) aged 24–44 years received Essure® for unilateral or bilateral hydrosalpinges. In most studies, Essure® was used for management of hydrosalpinges because of previous pelvic adhesions secondary to abdominopelvic surgery, endometriosis, Crohn's disease, pelvic inflammatory disease and tuberculosis. Essure® was chosen as a treatment modality for hydrosalpinx to avoid operative morbidity. In two studies, presence of pelvic adhesions and the associated condition was not a prerequisite for women with hydrosalpinges to undergo Essure® insertion.[31, 35] Outpatient setting with local anaesthetic use in the form of paracervical block and/or intravenous sedation was the most commonly used method of anaesthesia. In some women no analgesia or sedation was needed.[32, 36] In two case series, Essure® insertions were performed under general anaesthesia.[30, 37]

Procedure-related outcomes

Essure® was successfully placed in 96.5% of women (95% CI 91.1–98.9%) and 96.4% of the fallopian tubes (95% CI 91.7–98.7%; Table 3). The placement was successful on first attempt in all women except one where a second attempt was successful 1 month after the first unsuccessful attempt.[34] Tubal occlusion after Essure® insertion was confirmed in 98.1% of women (95% CI 93.1–99.9%) and 98.3% of the fallopian tubes (95% CI 93.8–99.9%). This was assessed in most studies by hysterosalpingogram 3 months after insertion except in two where it was assessed by ultrasound scan.[21, 37] A small number of women were awaiting assessment[31] and in a further small group of women there is not a clear mention of confirmation of tubal occlusion.[30, 31] Two procedure-related complications were seen: creation of a false passage during the insertion of Essure® because of the severe uterocervical angulation secondary to pelvic adhesions[30] and a single case of pyosalpinx.[37]

Table 3. Procedure-related and in vitro fertilisation (IVF)-related outcomes
Study IDProcedure-related outcomesIVF-related outcomes
Successful placementTubal occlusion confirmedIVF-embryo transfersPregnancy rate/ETMiscarriage rate/pregnancyLive birth rate/ETPreterm birth rate/ETOngoing pregnancy +live birth rate/ET
Per womanPer tubePer womanPer tube      
  1. a

    Data missing.

  2. b

    Tubal occlusion not confirmed in the remaining patients.

  3. c

    Not applicable as no clinical pregnancy.

  4. d

    One woman was 3 months pregnant at time of publication.

  5. e

    Does not include the spontaneous pregnancy in one patient after insertion of Essure® before IVF-embryo transfer could be done.

  6. f

    Not applicable as no embryo transfer.

  7. g

    Successful placement in one woman on second attempt.

  8. h

    Confirmation by ultrasound.

  9. i

    Included one fetal death at 27 weeks of gestation.

  10. j

    Only data of patient on study arm, i.e. Essure®, included.

  11. k

    3/15 had confirmation by ultrasound.

  12. l

    Three women had donated oocytes but embryo transfer had not been done, two women had not yet undergone IVF, while one women spontaneously conceived before IVF.

Rosenfield et al.[13]1/11/11/11/11 in 1 woman1/10/11/11/11/1
Hitkari et al.[30]3/55/81/1b1/1b2 in 2 women0/2c 0/20/20/2
Galen[31]6/711/124/4b a 5 in 5 women4/50/4d3/50/54/5
Gil and Arias[32]26/26a26/26aaaaaaa
Omurtag et al.[33]1/11/11/11/10 in 1 womanefffff
Nichols and West[34]6/6g11/11g6/611/1110 in 6 women5/102/53/103/103/10
Galen et al.[35]20/2032/3219/2031/3221 in 18 women13/211/1312/213/2112/21
Mijatovic et al.[36]19/2026/2719/2026/2745 in 19 women18/456/1812/451/4512/45
Thébault et al.[37]13/1323/2313/13h23/23h34 in 13 women7/344/7i3/340/343/34
Inocencio et al.[38]1/11/11/11/11 in 1 woman1/10/11/10/11/1
Matorras et al.[39]15/15j24/2415/15k24/2421 in 9 womenl5/21l1/5l4/213/214/21
Total111/115 (96.5%)135/140 (96.4%)106/108 (98.1%)119/121 (98.3%)140 in 75 women54/140 (38.6%)14/54 (25.9%)39/140 (27.9%)7/140 (5%)40/140 (28.6%)

Fertility-related outcomes

A total of 140 embryo transfers were performed in 75 women with a median of one embryo transfer per woman (range 0–5; Table 3). Two spontaneous pregnancies after insertion of Essure® but before IVF were reported.[21, 33] As a result of these 140 embryo transfers, there were 54 pregnancies (pregnancy rate per embryo transfer of 38.6% with 95% CI 30.9–46.8%), 39 live births (live birth rate per embryo transfer of 27.9 with 95% CI 21.1–35.8%) and seven preterm births (preterm live birth rate per embryo transfer of 5.0% with 95% CI 2.3–10.1%). There was one ongoing pregnancy,[31] so the combined ongoing pregnancy and live birth rate per embryo transfer was 28.6% (95% CI 21.7–36.6%). Of the 54 pregnancies, 14 resulted in miscarriages (25.9% with 95% CI 16.0–39.0%).

Discussion

Main findings

This systematic review and pooled analysis provide a summary of the use of Essure® for management of hydrosalpinx before IVF since it was first reported in 2005.[13] Successful placement of Essure® was achieved in nearly 96% of women and tubal occlusion was confirmed in 98% of those where successful placement was achieved. IVF subsequent to Essure® resulted in a pregnancy rate of 39% per embryo transfer, live birth rate of 28% and a combined ongoing pregnancy and live birth rate of 29% per embryo transfer.

Most studies were small case series or single case reports and only one had a control. This study did show that the IVF pregnancy rates following Essure® were somewhat lower (not statistically significant) than in those women who had laparoscopic salpingectomy for management of their hydrosalpinx.[39] Furthermore, the two cohorts in the study were not similar as Essure® was inserted only in those women where laparoscopic salpingectomy was contraindicated because of previous surgery and adhesions and this cohort also had a significantly lower ovarian reserve before Essure® insertion.

Interpretation

To better understand the procedure-related and IVF-related outcomes in this pooled analysis, similar outcomes in two other contexts were considered. These include, first, previously published data on outcomes related to the use of Essure® for hysteroscopic sterilisation and second, data on use of laparoscopic or surgical techniques in management of hydrosalpinx before IVF.

Procedure-related outcomes

While the experience of using Essure® for hydrosalpinx has been relatively new and occasional, its use for sterilisation has been much more extensive. It was approved for use as a hysteroscopic sterilisation device in Europe in 2001 and in USA in 2002.[11, 12] The initial phase II and phase III trials conducted from 1998 to 2001 showed successful placement of Essure® in 88–92% of women.[40, 41] Subsequently several studies have shown an improvement in the placement rate to 95–98%.[11, 42, 43] Following successful placement of Essure®, tubal occlusion has been seen in 92–96% of women as assessed by hysterosalpingogram 3 months after the procedure.[40, 41] These reported procedure-related outcomes are comparable to those in our analyses.

Sedation and analgesia for the procedure is mainly by use of paracervical block, nonsteroidal anti-inflammatory drugs and intravenous sedation, which allows insertion of Essure® to be generally performed in an outpatient or day-care setting.[40, 41] Patients report high satisfaction with the procedure[40-42] and it is more cost-effective than laparoscopy (which necessitates general anaesthesia) for sterilisation.[44, 45] In this pooled analysis, local anaesthesia in the form of paracervical block and/or intravenous sedation was most commonly used for the procedure except in two case series where general anaesthesia was used.[30, 37]

Commonly reported complications include expulsion and perforation in 1–3% of women, mainly when Essure® has either been placed too proximally or it has been difficult to visualise the ostia of the fallopian tube.[40, 41] The complication rate in our analysis was very low. Although no perforations were reported, a false passage was created in one woman during the insertion of Essure® because of the severe uterocervical angulation secondary to pelvic adhesions.[30]

IVF-related outcomes

In the UK, nearly 2% of all the babies born have been conceived through IVF treatment and the live birth rate per embryo transfer after IVF is around 25%.[46] These outcomes however vary by patient, diagnosis and IVF procedure-related factors.[47] Presence of hydrosalpinx reduces the pregnancy rates and live birth rates of infertile women undergoing IVF by 50% in comparison to infertile women without hydrosalpinx.[6, 7] In a meta-analysis of 14 studies, the pregnancy rate per embryo transfer of infertile women without hydrosalpinx was 31.2% compared with a rate of 19.7% in those with hydrosalpinx.[6] The corresponding figures for live birth rates per embryo transfer were 23.4 and 13.4%, respectively.[6]

A recent Cochrane review showed that laparoscopic salpingectomy/proximal tubal occlusion for hydrosalpinges before IVF increased the odds of pregnancy by over twofold with a pregnancy rate of 37% in the treatment arm (laparoscopic salpingectomy/proximal tubal occlusion) and 19% in the control arm (no intervention) although the rates were per woman and not per embryo transfer.[8] Data were insufficient to draw a conclusion on the effect of laparoscopic salpingectomy/proximal tubal occlusion for hydrosalpinges before IVF on live birth rates. The pregnancy rates and live birth rates of our pooled analyses compare favourably with these published outcomes.

Of particular interest is the observation of spontaneous pregnancy in two women with hydrosalpinx after insertion of Essure® but before IVF.[21, 33] In one of these women the hydrosalpinx was unilateral[33] whereas for the other woman it is not known if the hydrosalpinx was unilateral or bilateral.[21] Based on the proposed mechanisms of the effect of hydrosalpinx, including a direct embryotoxic effect, a decrease in endometrial receptivity and a mechanical flushing effect,[48-53] it is possible that in some situations occlusion of a unilateral hydrosalpinx would be sufficient to create an environment where natural conception can take place.

Importantly, no adverse effects on the fetus have been reported following successful pregnancies after the use of Essure® in those with hydrosalpinx.[13, 21, 31, 34-37] or in those with unintended pregnancies after use of Essure® for sterilisation.[54-56] In addition, the nickel titanium alloy used for the coil has shown no cytotoxic or genotoxic effects on animal models.[57] Polyethylene terephthalate fibres, which run inside the coil, have been used in various grafts, sutures and implants and have been found to be safe.

Strengths and limitations

This systematic review provides an up to date evidence-based summary of the role of Essure® in those with hydrosalpinx by searching published and grey literature including major relevant scientific meetings and having no language restriction minimised publication bias. It is possible that women in whom Essure® did not lead to benefit may not have been presented at scientific meetings or their data may not have been published. We also recognise that the majority of our studies are case reports and series and so the quality of evidence is not strong. For these reasons, we conclude that Essure® may be a safe and effective option and needs further study. Ultimately, controlled clinical trials with randomisation would provide the most robust evidence. The results of the ongoing DESH (Dutch Essure® versus Salpingectomy for Hydrosalpinx) Trial would be informative.[29] In this trial women who have hydrosalpinx and are suitable for laparoscopic surgery are randomised to receive hysteroscopic Essure® insertion and laparoscopic salpingectomy before IVF.

Conclusion

This review has the following implications for practice and research. It demonstrates that Essure® seems to be an effective option for management of hydrosalpinx in women before IVF where other operative treatment options are limited by the presence of pelvic adhesions. For this group of women with infertility, Essure® could indeed become the preferred alternative. However, for women with hydrosalpinx where presence of pelvic adhesions is not a complicating factor, Essure® is a nonsurgical option among a range of other surgical options like laparoscopic salpingectomy or proximal tubal occlusion. In the future, a controlled clinical trial, ideally with randomisation between Essure® and other established treatment options that measure IVF outcomes, adverse effects, patient experience and cost effectiveness would provide a more definitive answer to the role of Essure®.

Disclosure of interests

There are no financial disclosures.

Contribution to authorship

PA was responsible for conceiving, designing and co-ordinating the systematic review; designing search strategies and undertaking searches; searching grey literature; screening initial search and then filtering based on inclusion criteria; writing to authors of papers for additional information; data extraction; data analysis and interpretation; writing the manuscript and approving the final draft of the manuscript. RSA was responsible for screening initial search and then filtering based on inclusion criteria; data extraction; writing the manuscript and approving the final draft of the manuscript. DC was responsible for data analysis and interpretation; writing the manuscript and approving the final draft of the manuscript; providing a clinical perspective; and providing general advice on the systematic review.

Details of ethics approval

Approval was not needed as no human or animal participants were involved, and no medical records were accessed.

Funding

No funding was received.

Ancillary