Hysteroscopic hydrosalpinx occlusion with Essure device in

IVF atients when salpingectomy or laparoscopy is

contraindicated
ABSTRACT

Objective: To evaluate, in patients with hydrosalpinges, the effect on in vitro fertilization (IVF) outcome
of the insertion by hysteroscopy of an intratubal blocking device, in cases where laparoscopic
salpingectomy or laparoscopy was contraindicated.
Study design: A prospective interventional case series study was conducted in fifteen women with
unilateral (N = 6) or bilateral hydrosalpinges (N = 9) submitted for IVF. In all of them, laparoscopic
salpingectomy was contraindicated. Hysteroscopic insertion of the Essure intratubal device in a
consultation room setting was performed. IVF results were compared with those of women where
hydrosalpinx was treated by laparoscopic salpingectomy (48 women, 76 cycles).
Results: There were no complications during or immediately after the procedure in any of the patients.
There were four pregnancies from 16 embryo-transfers with own oocytes, one spontaneous pregnancy
after unilateral Essure insertion, and one pregnancy after oocyte donation. In one case the hydrosalpinx
grew and pelvic inflammatory disease developed 6 months after the insertion, requiring bilateral
adnexectomy. Although not of statistical significance, IVF pregnancy rates were somewhat lower than in
the laparoscopic salpingectomy group, which was attributed to the lower ovarian reserve before Essure
insertion.
Conclusion: The hysteroscopic insertion of the Essure intratubal device prior to IVF is a reasonable option
in cases where laparoscopic salpingectomy is contraindicated. Larger series are required to assess
pregnancy outcome.

1. Introduction
In tubal factor infertility due to bilateral hydrosalpinx, in vitro
fertilization (IVF) is the first option rather than attempting to
restore tubal function [1]. The hydrosalpinges, however, adversely
affect IVF outcomes, reducing the implantation rate and increasing
the risk of miscarriage [2,3]. A number of underlying pathogenic
mechanisms have been proposed: embryotoxic effects, mechanical
flushing and changes in endometrial receptivity [1,4]. Since the
hydrosalpinx fluid is in free communication with the uterine
cavity, any surgical intervention interrupting this communication
could improve the pregnancy rates [1,5,6]. Laparoscopic salpingectomy
has been demonstrated to be an effective option and has the advantages of removing the risk of pelvic
inflammatory disease
and ectopic pregnancies and of improving the accessibility of
ovarian tissue. It has drawbacks, however, including the invasiveness,
potential risks of surgical injury and anesthesia, and technical
difficulty in the case of adhesions [1]. Given this, a number of
alternative options have been proposed such as laparoscopic
proximal tubal occlusion [6], ultrasound-guided hydrosalpinx
aspiration [7], and Essure insertion [811].
The Essure device induces a benign local fibrous tissue response
that results in occlusion of the tubal lumen and encapsulation of the
device [12], and it is now widely used for tubal sterilization. Recently
Essure insertion has been proposed as a safe, effective and minimally
invasive alternative to salpingectomy in IVF patients with hydrosalpinges,
especially when laparoscopic treatment is contraindicated,
for example in cases with severe obesity, massive adhesions or
anesthesic risk [8,11,1315]. IVF outcome after Essure occlusion of
the hydrosalpinx has been reported to be similar to the general
population [9] or lower [15], although in none of the studies was a
control group presented, nor were IVF data presented in detail. The aim of this work is to report our
experience with the Essure
device in IVF patients with hydrosalpinx where laparoscopic
salpingectomy was contraindicated, and to compare IVF results
with those obtained in patients where hydrosalpinx was treated by
laparoscopic salpingectomy.
2. Materials and methods
The population under study consisted of 15 consecutive women
referred for IVF at our Reproduction Unit in whom an Essure device
(Conceptus Inc., Mountain View, CA) was inserted to occlude a
hydrosalpinx prior to IVF between 2005 and 2010. Throughout that
period, the first-line therapy in cases where IVF was indicated and
a hydrosalpinx was found was laparoscopic salpingectomy before
the IVF. During the same period, 48 women underwent laparoscopic
salpingectomy (salpingectomy group) and this population
is used as a control group. A second comparison group was
established including all the women with tubal factor infertility
undergoing IVF (n = 264) in the same period, in whom hydrosalpinx
had not been detected in the systematic pre-IVF
examination (tubal factor without hydrosalpinx group).
All the IVF cycles reported in the three populations involved
women under 40 years of age, undergoing IVF using their own
fresh oocytes. However, data concerning oocyte donation (OD)

cycles in the study population carried out in other centers are also
reported.
The infertility work-up of our patients before IVF included
sperm analysis, measurement of day 3 hormone levels, vaginal
ultrasound and hysterosalpingography (HSG) (except in severe
male factor cases in which ICSI was required). Tubal factor
infertility was defined as the lack of tubal patency evidenced by
HSG and/or laparoscopy, and hydrosalpinx as a distally occluded
tube that was pathologically dilated at HSG or laparoscopy. Since at
the HSG an image consistent with hydrosalpinx could correspond
to a true hydrosalpinx, or to a distally obstructed tube that
becomes dilated on the injection of radiographic contrast medium,
hydrosalpinx diagnosis in the present study was confirmed by
another complementary test (ultrasound and, if possible, laparoscopy).
Vaginal ultrasound was used in all cases, with a hydrosalpinx
being diagnosed when an elongated mass with incomplete
septum was observed. In the absence of contraindication,
laparoscopy was performed to confirm the presence of hydrosalpinges,
and affected tubes were surgically removed if technically
feasible. In cases where laparoscopic salpingectomy or even
laparoscopy was contraindicated because of surgical risk, Essure
insertion was offered to the patients.
In all cases there was a previous history of surgery, reporting
extensive pelvic adhesions (3 patients had laparotomy surgery and
another 3 had been operated on twice).
The hysteroscopic placement of the Essure microinsert was
performed in consultation rooms. Institutional board approval and
informed consent were obtained. No anesthetics were administered,
except for one case in which a paracervical block was
needed. The insertion was performed by hysteroscopists very
experienced in carrying out the procedure for tubal sterilization.
Diazepam (10 mg) and ibuprofen (600 mg) were given 30 min
before the insertion. Transvaginal ultrasound scans (n = 3) or HSG
(n = 12) were performed three months after the procedure to
evaluate whether the microinsert had been placed successfully. In
all cases the Essure was found to be adequately placed. In all cases
where HSG was performed, the proximal occlusion of the tube was
confirmed.
The IVF management used has been previously described
[16,17]. The present study has not received funding from the
manufacturer of the Essure device or any other commercial
organization.
For the statistical analysis Chi square, Fishers exact and
Students t tests were applied, following the standard criteria of
applicability. Binary logistic regression models were employed in
order to obtain adjusted estimates of the effect of the type of tubal
intervention.
3. Results
The Essure device was inserted while patients were on the IVF
waiting list (range: 311 months). In all 15 patients, it was possible
to carry out the insertion in the consultation room. The duration of
the procedure ranged between 5 and12 min and a mean number of
three coils were left protruding into the uterine cavity (range 14).
All the patients tolerated the procedure well and there were no
complications during or in the period immediately after the
insertion. Indications for Essure insertion were: extensive intraabdominal
adhesions of various etiologies: 9 cases of pelvic
endometriosis, 1 genital tract tuberculosis and 4 non-tuberculous
pelvic inflammatory disease (PID), and 2 massive adhesions of
unknown etiology. One of the patients with the diagnosis of
massive adhesion syndrome had a body mass index of 43.
In 6 cases the hydrosalpinx was unilateral and in 9 bilateral; in
unilateral cases devices were inserted only in the affected side. In
five cases, day 3 FSH levels were >10 mUI/mL.Of the 15 patients
who had the device inserted, two are still awaiting IVF. The
remaining 13 are listed in Table 1. There were 6 pregnancies, four
of them resulting from 19 IVF cycles started, corresponding to a
25% pregnancy rate per transfer (4/16). In one additional patient,
pregnancy was achieved with the devices inserted but after
bilateral adnexectomy, through OD. Another woman conceived
spontaneously after unilateral Essure insertion (case 1). She was a
31-year-old patient with a 5-year history of infertility, and became
pregnant 3 months after unilateral insertion of the Essure. Her
indication for Essure insertion was frozen pelvis, secondary to PID

and unilateral hydrosalpinx. The spontaneous pregnancy was a

twin gestation, but ended in a miscarriage at the seventh week.
Case 3 was referred due to stage IV endometriosis and bilateral
tubal factor, with extensive firm adhesions precluding visualization
of the tubes. Vaginal ultrasound showed bilateral hydrosalpinx,
and hysteroscopic blockage was indicated. Three months
after the insertion of the device ovarian stimulation for IVF was
started, but was cancelled because of a monofollicular response
and OD was recommended. She started hormone replacement
therapy in another center. The day before the embryo transfer of
the first OD cycle (six months after the Essure insertion), she was
admitted due to severe PID. Despite pharmacological treatment,
emergency surgery was needed and a bilateral adnexectomy was
performed by laparotomy without intraoperative complications.
Five months after the bilateral adnexectomy, frozen embryo
transfer was performed but did not lead to pregnancy. The second
OD cycle (15 months after Essure insertion) resulted in a single
pregnancy. She was admitted in week 30 because of premature
membrane rupture and preterm labor. A healthy preterm infant
was delivered by cesarean section due to breech presentation.
In case 6 a twin pregnancy occurred in the IVF cycle. Premature
rupture of membranes occurred at 31 weeks gestation: at 34 weeks
labor started, and a cesarean section was performed because of the
breech presentation of the first fetus. Two healthy infants were
delivered, and there were no postpartum complications.
Case 7 was referred due to three years of infertility and severe
pelvic and ureteral endometriosis. After the second oocyte
retrieval she was admitted due to a hemoperitoneum attributed
to ovarian bleeding, and recovered well with expectant management.
Pregnancy was not achieved.
Case 9 was a 39-year-old patient with two previous unsuccessful
IVF cycles. Laparoscopic treatment of bilateral hydrosalpinges was contraindicated due to frozen pelvis
diagnosed at previous
surgery. Following the IVF, 11 months after the bilateral placement
of intratubal device, a triplet pregnancy occurred. She was
admitted to an obstetric ward at 31 weeks with preterm labor.
The clinical evolution was satisfactory after tocolytic treatment. At 34 weeks the patient had a cesarean
delivery of three healthy
infants, and there were no postpartum complications.
In case 10, a right hydrosalpinx was diagnosed during the
infertility work-up. She started her first IVF cycle, without success,
8 months after the unilateral insertion of the device. She became
pregnant in the second IVF ET cycle but had a spontaneous
miscarriage 7 weeks after embryo transfer.
Case 13 became pregnant on the first IVF cycle and had a normal
pregnancy and a term vaginal delivery induced after third
trimester sonographic diagnosis of meningocele. The infant
underwent surgery without complications and had no neurological
sequelae. Serial ultrasound examinations performed during
pregnancy, starting at the gestational age of 5 weeks, showed no
space conflict between the intratubal device and the gestational
sac (Fig. 1).
Regarding IVF, the proportion of cases with higher than normal
FSH levels was much greater in the Essure group than in either the
salpingectomy (p < 0.05) or non hydrosalpinx tubal factor
groups (p < 0.01), as was the frequency of advanced endometriosis
(p < 0.01) (Tables 2 and 3). Also, the number of oocytes, fertilized
oocytes and transferred embryos obtained was lower. Although
the pregnancy rate was somewhat lower in the Essure group, the
differences were not statistically significant (Table 3).
A multivariate analysis was performed concerning first IVF
cycles, showing that age, basal FSH > 10 mUI/mL and advanced
endometriosis were inversely related to achieving pregnancy, whereas the number of obtained and
fertilized oocytes was
directly related to pregnancy. The type of tubal surgery (Essure,
salpingectomy) was not an independent prognostic factor,
although the analysis was hampered by the low number of
cases. When logistic regression analysis was applied to assess the
effect of the type of tubal intervention (Essure or salpingectomy)
on the probability of pregnancy, adjusted for the number of
retrieved oocytes, the odds ratio of pregnancy for the Essure
method was 1.08 (CI = 0.186.41). When the analysis was
restricted to cases with fewer than 10 oocytes, similar results
were obtained: Essure insertion (Table 3) had an OR of 1.17

(CI = 0.177.85).
4. Comments
Hydrosalpinges impair IVF results, presumably due to the
detrimental effect of the hydosalpinx fluid on implantation [14].
Laparoscopic salpingectomy of the hydrosalpinx has been shown
to be a very effective way to avoid this problem, and IVF pregnancy
rates after this procedure are very similar to those in the general
population [1]. Laparoscopic salpingectomy, however, involves,
apart from a non-negligible cost, undergoing a surgical procedure
with inherent risks which are higher in cases with extensive firm
adhesions, as well as in other cases with certain associated
conditions, such as morbid obesity. It has also been suggested that
laparoscopic salpingectomy could impair the ovarian response to
gonadotropins [18]. A number of alternative options have
therefore been proposed such as laparoscopic proximal tubal
occlusion [6], ultrasound-guided hydrosalpinx aspiration [7], and
Essure insertion [8,10,11,19,20].
The Essure system is based on a hysteroscopically placed spring
employed as an alternative to laparoscopic tubal occlusion for
permanent sterilization. The device itself not only is spaceoccupying
but also incites a local inflammatory response that
results in fibrosis of the tubal lumen. Insertion of the device can be
carried out in an outpatient setting and is well tolerated with no or
local anaesthesia. Concerning feasibility, in our series intratubal
device insertion was uneventful in all cases, with acceptable
tolerance in spite of the lack of general anesthesia. Moreover, there
was no complication associated with the hysteroscopic procedures,
and patient satisfaction with the procedure was high, in
agreement with previous reports [10,11].
One patient who had long standing infertility with unilateral
hydrosalpinx became pregnant spontaneously shortly after
intratubal device insertion. A very similar case has been reported
previously [19]. In our opinion, these cases support the hypothesis
of an adverse effect of hydrosalpinges on implantation, and suggest
that there is a beneficial effect from their occlusion, in both natural
and assisted conception.
Previous work on IVF in Essure-occluded hydrosalpinx
patients [9,15] reported results similar to the normal population
[9] or somewhat lower [15]. IVF data were not reported [9,15],
however, and the control group was absent [15] or the register of
Society for Assisted Reproduction Technologies was used as
control group [15].
The analysis of our IVF results was hampered by the small
number of cases in our population as well as by the nonrandomized
design of our study. Pregnancy rates were somewhat
lower, presumably as a consequence of the significantly
fewer oocytes obtained from Essure carriers. In our opinion the
number of oocytes recovered being so much smaller should be
attributed not to an adverse effect of the device, but rather to the
previously existing poor prognosis factors, such as the high rates
of elevated levels of FSH and advanced endometriosis among
Essure carriers already before the insertion. Thus, our experience
indicates that the insertion of the intratubal device is a good option before IVF in cases where
laparoscopic salpingectomy is
contraindicated.
The potential relationship between the Essure metallic coils
protruding into the uterine cavity and obstetric complications is an
important theoretical concern. To date, a total of 31 cases of
pregnancies in IVF patients carrying an Essure device have been
reported, with 25 live births [9,11,15], one preterm twin delivery
[8], four miscarriages [15] and one case of immature delivery with
chorioamnionitis [11]. In this last case, no relationship between the
Essure device and the obstetric outcome could be established. We
report a further six pregnancies in Essure carriers, including one
term vaginal delivery and three preterm cesarean section
deliveries. Two of the preterm deliveries occurred in multiples
pregnancies (one triplet) following premature rupture of the
membranes. These obstetric complications seemed to be related to
the multiple pregnancies. In both cases, the cesarean sections were
performed without complications and the Essure device was
apparently completely encapsulated and did not extend into the
uterine cavity. There was also one case of spontaneous preterm
rupture of membranes in a singleton pregnancy, where the role of
the tubal device remains uncertain. Further, there was a case of a

meningocele which seems to have had no relation at all to the

intratubal device. We also report two cases of miscarriage.
In our series there was one patient who had a complication in
which association with the Essure device could not be ruled out.
She was a woman in whom PID occurred during substitutive
hormonal therapy for OD (6 months after Essure insertion). Her
condition did not respond to antibiotic therapy and bilateral
adnexectomy was required. It could be speculated that the
proximal occlusion of the tube could be associated with
accumulation of fluid inside the tube, not allowing tubal fluid to
egress into the uterine cavity, and perhaps thereby favoring a
subsequent infection.
It should be remembered, however, that Essure was not
employed as an alternative to easy salpingectomies, but rather
for cases in which not even laparoscopic tubal occlusion was
feasible. Hypothetically, laparoscopic occlusion could also cause an
increase in the accumulation of fluid in the fallopian tubes. In our
opinion, laparoscopy should be the first step for diagnosing, and
where possible treating, the hydrosalpinges. Indeed, it offers the
advantage of providing a more accurate diagnosis of other
conditions (such as tuberculosis in our series). We conclude that
when laparoscopic surgical treatment of a hydrosalpinx is not
feasible, or when laparoscopy is not recommended, hysteroscopic
insertion of the Essure device is a reasonable option, which,
moreover, provides information concerning the status of the
uterine cavity.