- Email: [email protected]
Ovarian Response and in Vitro Fertilization Outcomes After Salpingectomy: Does Salpingectomy Indication Matter?
Accepted Manuscript Ovarian response and in vitro fertilization outcomes after salpingectomy: does salpingectomy indication matter? Nigel Pereira, M.D., Katherine P. Pryor, M.D., M.S., Anna Voskuilen-Gonzalez, M.D., Jovana P. Lekovich, M.D., Rony T. Elias, M.D., Steven D. Spandorfer, M.D., Zev Rosenwaks, M.D. PII:
S1553-4650(17)30002-X
DOI:
10.1016/j.jmig.2016.12.023
Reference:
JMIG 3028
To appear in:
The Journal of Minimally Invasive Gynecology
Received Date: 30 September 2016 Revised Date:
24 December 2016
Accepted Date: 31 December 2016
Please cite this article as: Pereira N, Pryor KP, Voskuilen-Gonzalez A, Lekovich JP, Elias RT, Spandorfer SD, Rosenwaks Z, Ovarian response and in vitro fertilization outcomes after salpingectomy: does salpingectomy indication matter?, The Journal of Minimally Invasive Gynecology (2017), doi: 10.1016/j.jmig.2016.12.023. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Pereira 1 Ovarian response and in vitro fertilization outcomes after salpingectomy: does
2
salpingectomy indication matter?
3
Nigel Pereira, M.D.,1 Katherine P. Pryor, M.D., M.S.,2 Anna Voskuilen-Gonzalez, M.D.,1 Jovana
4
P. Lekovich, M.D.,1 Rony T. Elias, M.D.,1 Steven D. Spandorfer, M.D.,1 Zev Rosenwaks, M.D.1
5
1
6
Medicine, New York, New York.
7
2
8
York
SC
The Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell
M AN U
Department of Obstetrics and Gynecology, Weill Cornell Medical College, New York, New
9 10
RI PT
1
Financial Disclosures: None for all authors
11
Corresponding author: Nigel Pereira, M.D., The Ronald O. Perelman and Claudia Cohen Center
13
for Reproductive Medicine, Weill Cornell Medicine, 1305 York Ave., New York, New York
14
10021 (E-mail: [email protected]).
EP
16
Short title: Ovarian response and IVF outcomes stratified by salpingectomy indication
AC C
15
TE D
12
ACCEPTED MANUSCRIPT Pereira 2 Précis
18
Patients undergoing in vitro fertilization after salpingectomy for ectopic pregnancy, unilateral
19
hydrosalpinx, bilateral hydrosalpinges, or other indications have similar oocyte yield,
20
fertilization, implantation and clinical pregnancy rates.
AC C
EP
TE D
M AN U
SC
RI PT
17
ACCEPTED MANUSCRIPT Pereira 3 Abstract
22
Study Objective: To investigate whether the ovarian response and pregnancy outcomes of
23
patients undergoing in vitro fertilization (IVF) after salpingectomy is affected by the underlying
24
indication for salpingectomy.
25
Design: Retrospective cohort study.
26
Design Classification: II-3.
27
Setting: University-affiliated fertility center.
28
Patients: All patients <37 years of age undergoing IVF within 12 months of laparoscopic
29
salpingectomy. The underlying indication for laparoscopic salpingectomy in the study cohort
30
was either tubal ectopic pregnancy, unilateral or bilateral hydrosalpinx, or other (hematosalpinx
31
or pyosalpinx), as confirmed by histopathology.
32
Interventions: IVF and embryo transfer (ET).
33
Measurements and Main Results: Surgical characteristics, demographics, ovarian stimulation
34
parameters, total oocytes retrieved, fertilization rates, implantation rates, and clinical pregnancy
35
rates were compared among the salpingectomy groups. Age and time-matched patients
36
undergoing their first IVF-ET cycle for male factor infertility, with no prior history of
37
laparoscopy served as controls.
38
Results: Of the 996 laparoscopic procedures during the study period, 136 patients underwent
39
unilateral salpingectomy for the following indications – ectopic pregnancy: 39; unilateral
40
hydrosalpinx: 81; other: 16. Of these, 29 patients in the ectopic pregnancy group, 75 patients in
41
the unilateral hydrosalpinx group, and 10 patients in the ‘other’ group underwent subsequent
AC C
EP
TE D
M AN U
SC
RI PT
21
ACCEPTED MANUSCRIPT Pereira 4 IVF-ET. Thirty-one patients underwent both bilateral salpingectomy and subsequent IVF-ET.
43
There was no difference in the antral follicle counts before and after salpingectomy in all groups.
44
There was a statistical difference in the mean duration of ovarian stimulation in the
45
salpingectomy groups – ectopic pregnancy: 10.9 ±2.15 days; unilateral hydrosalpinx: 9.56 ±1.95
46
days; bilateral hydrosalpinx: 9.51 ±2.01 days; other: 9.89 ±2.20 days; control: 9.76 ±1.99 days.
47
Similar trends were noted for total gonadotropins administered when comparing the ectopic
48
pregnancy group (3375.9 ±931.0 IU) to the remaining groups (unilateral hydrosalpinx: 2841.3
49
±1160.9 IU; bilateral hydrosalpinx: 2519.3 ±1004.7 IU; other: 2808.6 ±990.1 IU; control: 2726.1
50
±1129.8 IU). There was no difference in the total number of oocytes retrieved, fertilization rate,
51
implantation rate or clinical pregnancy rate in the salpingectomy groups compared to controls.
52
Conclusions: Although our findings indicate that patients undergoing IVF after salpingectomy
53
for an ectopic pregnancy have a statistically longer duration of stimulation and require higher
54
gonadotropin doses than patients undergoing IVF after salpingectomy for other indications, these
55
differences are of limited clinical significance given that the total number of oocytes retrieved,
56
implantation rate, and clinical pregnancy rate among the different salpingectomy groups was
57
comparable to controls.
58
Key Words: salpingectomy; laparoscopy; reproductive surgery; ovarian reserve; ovarian
59
response; ectopic pregnancy; hydrosalpinx; in vitro fertilization
AC C
EP
TE D
M AN U
SC
RI PT
42
ACCEPTED MANUSCRIPT Pereira 5 60 61
Introduction Salpingectomy is a commonly performed gynecologic procedure that involves the surgical removal of one or both fallopian tubes. In reproductive-aged women, salpingectomy is
63
routinely performed for the management of tubal pathologies, especially ectopic pregnancy and
64
hydrosalpinx. Lately, it has also been utilized prophylactically in women undergoing
65
hysterectomy to reduce the risk of ovarian cancer (1-4). Recent studies have suggested that
66
salpingectomy may have deleterious effects on ovarian reserve in all women, as well as ovarian
67
responsiveness in women undergoing various fertility treatments (5-8). The mechanism
68
underlying these outcomes is the theoretical disruption of collateral blood supply to the ovary
69
(5,7). However, other studies have refuted any detrimental impact of salpingectomy on ovarian
70
reserve and response (9-14).
SC
M AN U
These contrasting results in the medical literature (5-14) can be attributed to marked
TE D
71
RI PT
62
variations in study methodology (retrospective vs. randomized controlled trials vs. meta-
73
analysis), sample sizes, patient population (reproductive-aged women vs. women undergoing
74
hysterectomy), markers of ovarian reserve (antral follicle counts vs. follicle stimulating hormone
75
levels vs. anti-müllerian hormone levels), and markers of ovarian responsiveness (oocytes
76
retrieved vs. gonadotropins administered). Furthermore, several studies investigating the effects
77
of salpingectomy on ovarian reserve and response have included other concomitant surgical
78
procedures or diagnoses that may confound results. Thus, the objective of the current study is to
79
investigate whether the ovarian response and pregnancy outcomes of patients undergoing in vitro
80
fertilization (IVF) after salpingectomy is affected by the underlying indication for
81
salpingectomy.
AC C
EP
72
ACCEPTED MANUSCRIPT Pereira 6 82
Materials and Methods
83
Inclusion and Exclusion Criteria All patients <37 years of age undergoing ovarian stimulation for in vitro fertilization
RI PT
84
(IVF) after laparoscopic surgery at our center between January 2008 and June 2015 were
86
assessed for potential inclusion. Of this initial cohort of patients, only those undergoing IVF
87
within 12 months of laparoscopic salpingectomy were included. By study design, patients
88
undergoing salpingectomy via a non-laparoscopic approach were excluded. In addition, patients
89
undergoing concomitant procedures such as hysterectomy or oophorectomy were also excluded.
90
All patients undergoing IVF had normal baseline ovarian reserve as highlighted by the following
91
criteria: cycle day-2/3 follicle stimulating hormone (FSH) level <12 mIU/mL, cycle day-2/3 anti-
92
müllerian hormone (AMH) level >1 ng/mL, and an antral follicle count (AFC) of 8-15 (15). The
93
institutional review board at Weill Cornell Medical College approved our study protocol.
94
Surgical Methods
M AN U
TE D
95
SC
85
The underlying indication for laparoscopic salpingectomy in the study cohort was either tubal ectopic pregnancy, unilateral or bilateral hydrosalpinx and ‘other’. An ectopic pregnancy
97
was diagnosed with standard transvaginal ultrasonographic criteria in the setting of abnormally
98
rising beta human chorionic gonadotropin (βhCG) levels (16,17). In general, patients with an
99
ectopic pregnancy >3.5 cm, βhCG level >5000 mIU/mL, fetal cardiac activity or abdominal pain
AC C
EP
96
100
underwent unilateral laparoscopic salpingectomy with histopathologic confirmation (18). All
101
patients with evidence of hydrosalpinx on transvaginal ultrasonography or
102
hysterosalpingography underwent unilateral or bilateral laparoscopic salpingectomy prior to IVF
103
(19,20). Patients were considered to have a hydrosalpinx if the ultrasonographic and
ACCEPTED MANUSCRIPT Pereira 7 104
histopathologic diagnoses were consistent. Patients with histopathologic evidence of
105
hematosalpinx or pyosalpinx after laparoscopic salpingectomy were considered to have ‘other.’ All salpingectomies were performed by 1 of 4 surgeons using a standard laparoscopic
RI PT
106
technique (18). Briefly, salpingectomies were performed by cauterizing the mesosalpinx along
108
the entire length of the fallopian tube, using either a 5-mm Harmonic® scalpel (Ethicon Endo-
109
Surgery, J&J Medical Ltd, Cincinnati, USA) or LigaSureTM device (Covidien, Dublin, Ireland).
110
Cauterization was performed beginning at the fimbriated end running to the cornual end, with
111
care taken to cauterize as closely as possible to the fallopian tube to minimize damage to the
112
tubal vascular supply. The proximal end of the fallopian tube was transected. The transected
113
fallopian tube was removed through a 12-mm port using an endoscopic bag.
114
Clinical Protocols
M AN U
Ovarian stimulation was performed with gonadotropins (Follistim, Merck, Kenilworth,
TE D
115
SC
107
NJ, USA; Gonal-F, EMD-Serono Inc., Rockland, MA, USA; and Menopur, Ferring
117
Pharmaceuticals Inc, Parsippany, NJ, USA) to maximize follicular response while minimizing
118
the risk of ovarian hyperstimulation syndrome (21). Gonadotropin doses were based on patient
119
age, weight, antral follicle count (AFC), and serum anti-müllerian hormone (AMH) levels.
120
Ovulation was suppressed using daily injections of 0.25 mg Ganirelix Acetate (Merck,
121
Kenilworth, NJ, USA). hCG was used as the ovulatory trigger, which was given when the two
122
lead follicles attained a mean diameter >17 mm. Oocyte retrieval was performed under conscious
123
sedation using a 30 cm 16 G oocyte aspiration needle (Cook® Medical, Bloomington, IN, USA)
124
using transvaginal ultrasound guidance approximately 35–37 hours after hCG administration.
AC C
EP
116
ACCEPTED MANUSCRIPT Pereira 8 125
The retrieved oocytes were fertilized using conventional in vitro insemination or intracytoplasmic sperm injection (ICSI) depending on the couple’s reproductive history. Oocytes
127
were examined 14-18 hours after conventional insemination or ICSI to assess the fertilization
128
rate i.e., number of 2-pronuclear embryos out of the number of mature oocytes inseminated or
129
injected (22). The resulting embryos were cultured in in-house media and then graded at the
130
cleavage stage using the Veeck criteria (23). All fresh ETs were performed on day 3 (cleavage-
131
stage) with Wallace catheters (Smiths Medical Inc., Norwell, MA, USA) at approximately 1 cm
132
less than the uterine depth identified at prior trial transfer (24). Ultrasound guidance was utilized
133
only when ETs were deemed difficult based on the prior trial transfer (24). Surplus embryos
134
were cultured to the blastocyst-stage and the top-quality blastocysts were cryopreserved by
135
vitrification (25). No patients underwent pre-implantation genetic screening of embryos in the
136
current study.
137
Variables Recorded
SC
M AN U
TE D
138
RI PT
126
Surgical characteristics recorded were the laterality of the excised fallopian tube (right vs. left), surgeon performing the laparoscopic salpingectomy, the laparoscopic instrument utilized
140
(Harmonic® scalpel or LigaSureTM device) and the final histopathologic diagnosis. The time
141
between the laparoscopic salpingectomy and initiation of ovarian stimulation was also recorded.
142
Baseline demographic characteristics recorded for all patients included age and body mass index
143
(BMI, kg/m2). Cycle day 2/3 AFC and AMH levels (ng/mL) were measured in all patients prior
144
to laparoscopy. The baseline IVF parameters of all patients prior to ovarian stimulation were also
145
noted, which included AFC, cycle day 2/3 follicle stimulating hormone (FSH, mIU/mL) level,
146
cycle day 2/3 estradiol (E2, pg/mL) level, and infertility diagnosis. Ovarian stimulation
147
parameters assessed were the duration of ovarian stimulation (days), total gonadotropins
AC C
EP
139
ACCEPTED MANUSCRIPT Pereira 9 administered (IU), and total oocytes retrieved. Finally, the following IVF outcomes were also
149
noted: fertilization rate (%), mean number of embryos transferred, implantation rate defined as
150
the mean number of sacs seen on ultrasonography divided by the number of embryos transferred
151
for each patient, and clinical pregnancy rate defined as the number of intrauterine gestations with
152
fetal cardiac activity per IVF cycle. The number of surplus embryos cryopreserved at the
153
blastocyst-stage was recorded as well.
154
Statistical Analyses
SC
Continuous variables were assessed for normality by the Shapiro-Wilk’s test and all were
M AN U
155
RI PT
148
found to be normally distributed; all continuous variables are expressed as mean ± standard
157
deviation (SD). Categorical variables were expressed as number of cases (n) and percentage of
158
occurrence (%). Patients undergoing salpingectomy were stratified into 4 groups based on
159
histopathologic diagnosis i.e., ectopic pregnancy, unilateral hydrosalpinx, bilateral hydrosalpinx
160
and other. Age and time-matched patients undergoing their first IVF-ET cycle for male factor
161
infertility, with no prior history of laparoscopy served as controls. Baseline demographics,
162
baseline IVF parameters and the time between laparoscopic salpingectomy and initiation of
163
ovarian stimulation were compared between groupings of histopathologic diagnosis and controls
164
by using analysis of variance (ANOVA) tests. Chi-square (χ2) tests were used to compare
165
fertilization, implantation and clinical pregnancy rates between groupings of histopathologic
166
diagnosis and controls. Multiple linear regression models were used to investigate the
167
relationship between 1) salpingectomy indication and ovarian stimulation duration and 2)
168
salpingectomy indication and total gonadotropins administered, controlling for age, BMI, and
169
time elapsed between salpingectomy indication and ovarian stimulation. Also, in order to address
170
the potential issue of surgeon variability in performing the laparoscopic salpingectomy (i.e.,
AC C
EP
TE D
156
ACCEPTED MANUSCRIPT Pereira 10 potential clustering of outcomes within surgeon), robust standard errors were computed by using
172
generalized estimating equations (GEE) modeling, specifically clustering on performing surgeon.
173
In a previous investigation, we reported a gonadotropin dosage difference of 705 IU in patients
174
undergoing IVF after unilateral salpingectomy i.e., 2124 IU pre-salpingectomy and 3126 IU
175
post-salpingectomy (18). Based on these findings, a sample size of 27 patients per group was
176
estimated assuming an alpha level of 5% and a power of 80%. All statistical analyses were
177
performed using STATA version 14 (StataCorp, College Station, TX). Statistical significance
178
was set at P<0.05.
179
Results
M AN U
SC
RI PT
171
Nine hundred ninety-six patients underwent laparoscopy during the study period: 136
181
(13.7%) unilateral salpingectomy, 31 (3.11%) bilateral salpingectomy, and 829 (83.2%) other
182
laparoscopic procedures. Among patients undergoing unilateral salpingectomy, 39 (28.7%)
183
patients had an ectopic pregnancy, 81 (59.6%) patients had a hydrosalpinx, while the remaining
184
16 (11.7%) had other histopathologic diagnoses. Twenty-nine out of the 39 patients in the
185
ectopic pregnancy group, 75 out of the 81 patients in the hydrosalpinx group, and 10 out of the
186
16 patients in the ‘other’ group underwent subsequent ovarian stimulation. Of note, the final
187
histopathologic diagnosis in the other group was hematosalpinx (8 patients) and pyosalpinx (1
188
patient). The tenth patient underwent unilateral salpingectomy for spontaneous torsion of the
189
fallopian tube. All 31 patients in the bilateral salpingectomy group underwent salpingectomy and
190
subsequent ovarian stimulation. Figure 1 summarizes the selection of the study cohort.
191 192
AC C
EP
TE D
180
Table 1 compares the surgical characteristics of all patients undergoing salpingectomy stratified by salpingectomy indication. No difference was noted in the laterality of fallopian tube
ACCEPTED MANUSCRIPT Pereira 11 removed (right vs. left), type of surgical instrument used, or the time elapsed between
194
salpingectomy and subsequent ovarian stimulation. Of note, patients in the ectopic pregnancy,
195
unilateral hydrosalpinx and ‘other’ groups had underlying male factor or unexplained infertility;
196
patients in the bilateral hydrosalpinx group were considered to have tubal infertility. As evident
197
from table 2, no differences were noted in the baseline demographics of patients undergoing
198
salpingectomy. Furthermore, the AFC and AMH levels of all salpingectomy groups were similar,
199
indicating that the groups were comparable for ovarian reserve at baseline. Table 2 also shows
200
that the AFC, FSH and E2 levels of the salpingectomy groups were similar to controls even after
201
laparoscopy, suggesting no acute changes in ovarian reserve markers.
M AN U
SC
RI PT
193
Figure 2 (panel A) compares the duration of ovarian stimulation in patients after
203
salpingectomy. There was a statistically significant difference in the mean duration of ovarian
204
stimulation in the ectopic pregnancy group (10.9 ±2.15 days) compared to the unilateral
205
hydrosalpinx (9.56 ±1.95 days), bilateral hydrosalpinx (9.51 ±2.01 days), other (9.89 ±2.20
206
days), and control (9.76 ±1.99 days) groups (P=0.01). As seen in figure 2 (panel B), similar
207
trends were noted for the total gonadotropins administered when comparing the ectopic
208
pregnancy group (3375.9 ±931.0 IU) to the remaining groups (unilateral hydrosalpinx: 2841.3
209
±1160.9 IU; bilateral hydrosalpinx: 2519.3 ±1004.7 IU; other: 2808.6 ±990.1 IU; control: 2726.1
210
±1129.8 IU; P<0.001). The duration of ovarian stimulation and dosage of gonadotropins was
211
+0.12 days longer (table 3) and +211.4 IU higher (table 4), respectively in the ectopic group
212
compared to all other salpingectomy groups, after multiple linear regression accounting for age,
213
BMI and the time elapsed between salpingectomy and IVF-ET. Figure 3 (panel A) shows no
214
difference in the total number of oocytes retrieved in the salpingectomy groups compared to
215
controls. Figure 3 (panel B) indicates that there were no differences in the fertilization rate.
AC C
EP
TE D
202
ACCEPTED MANUSCRIPT Pereira 12 Furthermore, Figure 3 (panel C) also reveals that the mean embryos transferred, mean
217
implantation rates and clinical pregnancy rates were comparable between the salpingectomy
218
groups. The mean number of top-quality embryos cryopreserved at the blastocyst-stage in the
219
different salpingectomy groups was comparable and as follows: 1.72 ±0.99 (ectopic pregnancy);
220
1.94 ±0.88 (unilateral salpingectomy); 1.95 ±1.03 (bilateral salpingectomy); 1.87 ±0.98 (other);
221
1.80 ±0.87 (control).
222
Discussion
SC
RI PT
216
The current retrospective study investigated whether ovarian response or pregnancy
224
outcomes in patients undergoing IVF-ET after salpingectomy is affected by the underlying
225
indication for salpingectomy. While our findings suggest that patients undergoing IVF after
226
salpingectomy for an ectopic pregnancy have a statistically longer duration of ovarian
227
stimulation and require statistically higher doses of gonadotropins than patients undergoing IVF
228
after salpingectomy for other indications, these differences are of limited clinical significance
229
given that the total number of oocytes retrieved, fertilization rate, implantation rate, and clinical
230
pregnancy rate among the different salpingectomy groups was similar to controls.
TE D
EP
231
M AN U
223
Whether or not salpingectomy impairs ovarian reserve remains controversial. Several conflicting studies have investigated the effect of salpingectomy on ovarian reserve by looking at
233
surrogate markers in subsequent ovarian stimulation cycles. In a prospective study of 29 patients
234
who had undergone unilateral salpingectomy for management of ectopic pregnancy, Lass et al.
235
(26) noted decreased follicular development and fewer oocytes retrieved from the ovary on the
236
operated side. Similarly, Gelbaya et al. (27) noted significantly increased baseline FSH levels,
237
decreased E2 levels at time of trigger, decreased follicular development and fewer oocytes
AC C
232
ACCEPTED MANUSCRIPT Pereira 13 retrieved following salpingectomy performed for hydrosalpinx. Xi et al. (12) showed that women
239
undergoing salpingectomy for ectopic pregnancy required higher initial and total doses of
240
gonadotropins during subsequent IVF cycles as compared to non-operated controls. In contrast,
241
Ye at al. (5) did not observe differences in gonadotropin dosing, but instead noted significantly
242
decreased AMH in women after bilateral salpingectomy. Finally, Chan et al. (28) noted
243
decreased AFC and ovarian stromal blood flow on the operated side. At first glance, the
244
aforementioned studies may suggest that salpingectomy impairs ovarian reserve; however, there
245
are several other prospective and retrospective studies that have found no differences in ovarian
246
reserve following unilateral or bilateral salpingectomy performed for the management of either
247
ectopic pregnancy or hydrosalpinx (9-14, 29-32). Supplemental table 1 summarizes all studies
248
evaluating ovarian reserve or responsiveness after salpingectomy for ectopic pregnancy or
249
hydrosalpinx. As evident from the table, the large majority of studies have shown no difference
250
in ovarian reserve or responsiveness after salpingectomy for either indication.
SC
M AN U
TE D
251
RI PT
238
In a recent meta-analysis of 25 studies that included 1935 patients, Fan and Ma (7) suggested that salpingectomy impaired ovarian function. Specifically, when comparing subjects
253
undergoing unilateral salpingectomy to controls, there was a statistically significant decrease in
254
the number of oocytes retrieved. When comparing subjects undergoing bilateral salpingectomy
255
to controls, longer gonadotropin stimulation and higher doses were required, baseline FSH was
256
increased, and the number of oocytes retrieved was decreased. No differences were observed
257
when comparing the unilateral and bilateral salpingectomy groups. As the authors noted, the
258
meta-analysis was complicated by significant heterogeneity of the studies, including variability
259
in salpingectomy technique, as well as the time elapsed between surgery and subsequent ovarian
AC C
EP
252
ACCEPTED MANUSCRIPT Pereira 14 260
stimulation. Notably, the meta-analysis did not stratify results by the indication for
261
salpingectomy. Salpingectomy has been postulated to decrease ovarian reserve through several
RI PT
262
mechanisms. One common argument is that ovarian blood flow can be compromised during
264
salpingectomy. Because the tubal branch of the uterine artery originates at the same point as the
265
ovarian branch of the uterine artery, damage to ovarian branch as it traverses the mesosalpinx, by
266
surgical disruption or thermal spread can decrease blood flow to the ovary (18,27). It is also
267
possible that certain changes in blood flow to the fallopian tubes and ovaries through the
268
mesosalpinx may be unique to ectopic pregnancies. In ectopic pregnancies, trophoblastic
269
invasion of the fallopian tube increases blood flow through the tubal branch of the uterine artery
270
and promotes development of collateral blood flow (35). In addition, tubal pregnancies have
271
been shown to affect the pulsatility of the ipsilateral uterine artery (36). Thus, trophoblastic
272
invasion of the fallopian tube can affect ovarian blood flow such that at the time of
273
salpingectomy, the ovary has increased vulnerability to abrupt changes in blood supply. In
274
contrast, hemodynamic studies using Doppler ultrasonography in patients with hydrosalpinges
275
have revealed that the blood flow to the endometrium and ovary may be impaired at baseline
276
(37,38). Thus, an abrupt decrease in blood flow to ovary may not occur after salpingectomy for
277
hydrosalpinges as compared to tubal ectopic pregnancies.
M AN U
TE D
EP
AC C
278
SC
263
Some surgeons have also argued that disruption of blood flow may occur only with poor
279
surgical technique (39). Indeed, surgical technique and the type of laparoscopic instrument used
280
can also mediate the association between salpingectomy and reduced ovarian responsiveness
281
(39). However, as previously described, we used either a 5-mm Harmonic® scalpel or
282
LigaSureTM device to cauterize the mesosalpinx as close as possible to the fallopian tube to
ACCEPTED MANUSCRIPT Pereira 15 minimize inadvertent damage to ovarian blood supply. Furthermore, these instruments minimize
284
lateral thermal spread and instrument heating compared to monopolar or conventional bipolar
285
electrocautery, thereby reducing unintended tissue damage (40, 41).
286
RI PT
283
Although disruption in ovarian blood flow and surgical technique may contribute to the pathogenesis of decreased ovarian reserve after unilateral or bilateral salpingectomy, it is
288
important to highlight the existence of well-designed studies that rebut such results. One such
289
study is by Morelli et al. (14), in which the authors compared blood ovarian reserve parameters
290
(AMH and FSH levels) and ultrasonographic ovarian reserve parameters (AFC, mean ovarian
291
diameters and peak systolic velocity) in 79 patients 3 months before and after total laparoscopic
292
hysterectomy, with or without salpingectomy. Patients who underwent total laparoscopic
293
hysterectomy with salpingectomy showed no differences in blood or ultrasonographic ovarian
294
reserve parameters compared to those who underwent laparoscopic hysterectomy without
295
salpingectomy. Furthermore, both groups had comparable operative times, blood loss,
296
postoperative hospital stay and postoperative return to normal activity. A subsequent randomized
297
controlled trial by Venturella et al. (42) compared 91 patients undergoing standard laparoscopic
298
salpingectomy to 95 patients undergoing laparoscopic salpingectomy with wide excision of the
299
mesosalpinx. The authors reported no differences in AMH level, FSH levels, AFC, and vascular,
300
flow and vascular-flow indices. Similar to the Morelli et al. study, no differences in peri-
301
operative or post-operative outcomes were noted.
M AN U
TE D
EP
AC C
302
SC
287
It is important to emphasize that an association, if any, between impaired ovarian
303
response after salpingectomy is multi-factorial, including publication biases. The main strength
304
of the current study is the demonstration of similar oocyte yield, fertilization rates, implantation
305
rates and clinical pregnancy rates in patients undergoing IVF-ET after laparoscopic
ACCEPTED MANUSCRIPT Pereira 16 salpingectomy, irrespective of the underlying indication for salpingectomy. In addition, the
307
current study also accounted for several surgical variables (laterality of the fallopian tube,
308
surgical instrumentation, surgeon performing the laparoscopy) and the time elapsed between
309
laparoscopic salpingectomy and IVF, which have not been evaluated by prior studies.
310
RI PT
306
Despite these strengths, our study was limited by its retrospective nature and the small sample sizes of the salpingectomy groups. While the sample sizes were estimated to find
312
differences in gonadotropin dosage, post-hoc calculations suggest an approximate sample of size
313
of 376 per salpingectomy group to detect a 10% difference in clinical pregnancy rates. It is
314
worthwhile to note the unusually small magnitude of variation in the means and standard
315
deviations of all parameters listed in table 2. While this lack of variation can be attributed to the
316
normal baseline ovarian reserve parameters of all patients, the small sample size may also be an
317
overall source for such lack of variation. Another shortcoming of our study is that post-
318
salpingectomy AMH levels were not measured; however, it is reassuring to note that post-
319
salpingectomy AFC, FSH and E2 levels were similar to controls. Finally, ovarian response was
320
not evaluated separately in the ipsilateral ovary and the contralateral ovary when unilateral
321
salpingectomies were performed. While we acknowledge that study’s overall generalizability is
322
limited due to the aforementioned limitations, we deem that these findings do merit further
323
investigation in a prospective setting.
M AN U
TE D
EP
AC C
324
SC
311
In conclusion, the current study demonstrates similar oocyte yield and pregnancy
325
outcomes in patients undergoing IVF after salpingectomy, irrespective of the indication for
326
salpingectomy. Although the overall sample size remains small, the study did account for several
327
surgical variables such as laterality of the fallopian tube, surgical instrument used, performing
328
surgeon as well as the time elapsed between salpingectomy and ovarian stimulation. Thus, our
ACCEPTED MANUSCRIPT Pereira 17 findings can be useful to reassure patients undergoing ovarian stimulation for IVF-ET after
330
unilateral or bilateral laparoscopic salpingectomy.
AC C
EP
TE D
M AN U
SC
RI PT
329
ACCEPTED MANUSCRIPT Pereira 18
333 334 335 336
1. American College of Obstetricians and Gynecologists. Salpingectomy for ovarian cancer prevention. Committee Opinion No. 620. Obstet Gynecol 2015;125:279–81.
RI PT
332
References
2. Falconer H, Yin L, Grönberg H, Altman D. Ovarian cancer risk after salpingectomy: a nationwide population-based study. J Natl Cancer Inst 2015;107(2).
3. McAlpine JN, Hanley GE, Woo MM, Tone AA, Rozenberg N, Swenerton KD et al.
SC
331
Opportunistic salpingectomy: uptake, risks, and complications of a regional initiative for
338
ovarian cancer prevention. Am J Obstet Gynecol 2014;210(5):471.e1-11.
339
M AN U
337
4. Yoon SH, Kim SN, Shim SH, Kang SB, Lee SJ. Bilateral salpingectomy can reduce the
340
risk of ovarian cancer in the general population: A meta-analysis. Eur J Cancer
341
2016;55:38-46.
342
5. Ye XP, Yang YZ, Sun XX. A retrospective analysis of the effect of salpingectomy on serum antiMüllerian hormone level and ovarian reserve. Am J Obstet Gynecol
344
2015;212(1):53.e1-10.
345
TE D
343
6. Grynnerup AG, Lindhard A, Sørensen S. Anti-Müllerian hormone levels in salpingectomized compared with nonsalpingectomized women with tubal factor
347
infertility and women with unexplained infertility. Acta Obstet Gynecol Scand
349 350 351
AC C
348
EP
346
2013;92(11):1297-303.
7. Fan M, Ma L. Effect of salpingectomy on ovarian response to hyperstimulation during in vitro fertilization: a meta-analysis. Fertil Steril 2016;106(2):322-329.e9.
8. Orvieto R, Saar-Ryss B, Morgante G, Gemer O, Anteby EY, Meltcer S. Does
352
salpingectomy affect the ipsilateral ovarian response to gonadotropin during in vitro
353
fertilization-embryo transfer cycles? Fertil Steril 2011;95(5):1842-4.
ACCEPTED MANUSCRIPT Pereira 19 354 355 356
9. Qin F, Du DF, Li XL. The Effect of Salpingectomy on Ovarian Reserve and Ovarian Function. Obstet Gynecol Surv 2016;71(6):369-76. 10. Yoon SH, Lee JY, Kim SN, Chung HW, Park SY, Lee C. Does salpingectomy have a deleterious impact on ovarian response in in vitro fertilization cycles? Fertil Steril 2016
358
Jun 20. pii: S0015-0282(16)61300-6.
RI PT
357
11. Lin YJ, Ou YC, Huang FJ, Lin PY, Kung FT, Lan KC. Ovarian response to
360
gonadotropins in patients with tubal factor infertility: salpingectomy versus
361
nonsalpingectomy. J Minim Invasive Gynecol 2013;20(5):637-41.
12. Xi W, Gong F, Tang Y, Zhang H, Lu G. Ovarian response to gonadotropins after
M AN U
362
SC
359
363
laparoscopic salpingectomy for ectopic pregnancy. Int J Gynaecol Obstet
364
2012;116(2):93-6.
13. Almog B, Wagman I, Bibi G, Raz Y, Azem F, Groutz A, et al. Effects of salpingectomy
366
on ovarian response in controlled ovarian hyperstimulation for in vitro fertilization: a
367
reappraisal. Fertil Steril 2011;95(8):2474-6.
TE D
365
14. Morelli M, Venturella R, Mocciaro R, Di Cello A, Rania E, Lico D, et al. Prophylactic
369
salpingectomy in premenopausal low-risk women for ovarian cancer: primum non
370
nocere. Gynecol Oncol 2013;129(3):448-51.
EP
368
15. Pereira N, Friedman C, Hutchinson AP, Lekovich JP, Elias RT, Rosenwaks Z. Increased
372
odds of live birth in fresh in vitro fertilization cycles with shorter ovarian stimulation.
373 374
375 376
AC C
371
Fertil Steril 2016; pii: S0015-0282(16)62875-3. doi: 10.1016/j.fertnstert.2016.09.044. [Epub ahead of print]
ACCEPTED MANUSCRIPT Pereira 20 377
16. Barnhart KT. Clinical practice. Ectopic pregnancy. N Engl J Med 2009;361(4):379-87.
378
17. Kirk E, Papageorghiou AT, Condous G, Tan L, Bora S, Bourne T. The diagnostic effectiveness of an initial transvaginal scan in detecting ectopic pregnancy. Hum Reprod
380
2007;22(11):2824-8.
RI PT
379
18. Pereira N, Gerber D, Gerber RS, Lekovich JP, Elias RT, Spandorfer SD, Rosenwaks Z.
382
Effect of Methotrexate or Salpingectomy for Ectopic Pregnancy on Subsequent In Vitro
383
Fertilization-Embryo Transfer Outcomes. J Minim Invasive Gynecol 2015;22(5):870-6.
SC
381
19. Patel MD, Acord DL, Young SW. Likelihood ratio of sonographic findings in
385
discriminating hydrosalpinx from other adnexal masses. AJR Am J Roentgenol
386
2006;186(4):1033-8.
387
M AN U
384
20. Johnson N, van Voorst S, Sowter MC, Strandell A, Mol BW. Surgical treatment for tubal disease in women due to undergo in vitro fertilization. Cochrane Database Syst Rev
389
2010;(1):CD002125.
392 393 394 395 396 397 398
2014;1154:171-231.
22. Palermo GD, Neri QV, Rosenwaks Z. To ICSI or Not to ICSI. Semin Reprod Med
EP
391
21. Huang JY, Rosenwaks Z. Assisted reproductive techniques. Methods Mol Biol
2015;33(2):92-102.
23. Gosden LV. Oocyte retrieval and quality evaluation. Methods Mol Biol 2014;1154:343-
AC C
390
TE D
388
60.
24. Pereira N, Petrini AC, Lekovich JP, Schattman GL, Rosenwaks Z. Singleton pregnancy and perinatal outcomes after fresh versus frozen-thawed blastocyst transfers. Int J Gynaecol Obstet 2016;135(1):96-100.
ACCEPTED MANUSCRIPT Pereira 21 399 400 401
25. Kuwayama M. Highly efficient vitrification for cryopreservation of human oocytes and embryos: the Cryotop method. Theriogenology 2007;67(1):73-80. 26. Lass A, Ellenbogen A, Croucher C, Trew G, Margara R, Becattini C, Winston RM. Effect of salpingectomy on ovarian response to superovulation in an in vitro fertilization-
403
embryo transfer program. Fertil Steril 1998;70(6):1035-8.
404
RI PT
402
27. Gelbaya TA, Nardo LG, Fitzgerald CT, Horne G, Brison DR, Lieberman BA. Ovarian response to gonadotropins after laparoscopic salpingectomy or the division of fallopian
406
tubes for hydrosalpinges. Fertil Steril 2006;85(5):1464-8.
28. Chan CC, Ng EH, Li CF, Ho PC. Impaired ovarian blood flow and reduced antral follicle
M AN U
407
SC
405
408
count following laparoscopic salpingectomy for ectopic pregnancy. Hum Reprod
409
2003;18(10):2175-80.
410
29. Strandell A, Lindhard A, Waldenstrom U, Thorburn J, Janson PO, Hamberger L. Hydrosalpinx and IVF outcome: a prospective, randomized multicentre trial in
412
Scandinavia on salpingectomy prior to IVF. Hum Reprod 1999;14(11):2762-9.
413
TE D
411
30. Dar P, Sachs GS, Strassburger D, Bukovsky I, Arieli S. Ovarian function before and after salpingectomy in artificial reproductive technology patients. Hum Reprod
415
2000;15(1):142-4.
EP
414
31. Demir B, Bozdag G, Sengul O, Kahyaoglu I, Mumusoglu S, Zengin D. The impact of
417
unilateral salpingectomy on antral follicle count and ovarian response in ICSI cycles:
418 419
AC C
416
comparison of contralateral side. Gynecol Endocrinol 2016:1-4.
32. Tal J, Paltieli Y, Korobotchka R, Ziskind G, Eibschitz I, Ohel G. Ovarian response to
420
gonadotropin stimulation in repeated IVF cycles after unilateral salpingectomy. J Assist
421
Reprod Genet 2002;19(10):451-5.
ACCEPTED MANUSCRIPT Pereira 22 422
33. Kontoravdis A, Makrakis E, Pantos K, Botsis D, Deligeoroglou E, Creatsas G. Proximal
423
tubal occlusion and salpingectomy result in similar improvement in in vitro fertilization
424
outcome in patients with hydrosalpinx. Fertil Steril 2006;86(6):1642-9. 34. Verhulst G, Vandersteen N, van Steirteghem AC, Devroey P. Bilateral salpingectomy
RI PT
425 426
does not compromise ovarian stimulation in an in-vitro fertilization/embryo transfer
427
programme. Hum Reprod 1994;9(4):624-8.
35. Kirchler HC, Seebacher S, Alge AA, Muller-Holzner E, Fessler S, Kolle D. Early
SC
428
diagnosis of tubal pregnancy: changes in tubal blood flow evaluated by endovaginal color
430
Doppler sonography. Obstet Gynecol 1993;82(4 Pt 1):561-5
431 432
M AN U
429
36. Tekay A, Jouppila P. Color Doppler flow as an indicator of trophoblastic activity in tubal pregnancies detected by transvaginal ultrasound. Obstet Gynecol 1992;80(6):995-9. 37. El-Mazny A, Ramadan W, Kamel A, Gad-Allah S. Effect of hydrosalpinx on uterine and
434
ovarian hemodynamics in women with tubal factor infertility. Eur J Obstet Gynecol
435
Reprod Biol 2016;199:55-9.
436
TE D
433
38. Ng EH, Chan CC, Tang OS, Ho PC. Comparison of endometrial and subendometrial blood flows among patients with and without hydrosalpinx shown on scanning during in
438
vitro fertilization treatment. Fertil Steril 2006;85(2):333-8.
440
39. Dietl J, Wischhusen J. The postreproductive salpingectomy. Fertil Steril
AC C
439
EP
437
2014;101(3):e20.
441
40. Sutton PA, Awad S, Perkins AC, Lobo DN. Comparison of lateral thermal spread using
442
monopolar and bipolar diathermy, the Harmonic Scalpel and the Ligasure. Br J Surg
443
2010;97(3):428-33.
ACCEPTED MANUSCRIPT Pereira 23 444 445
41. Sutton C, Abbott J. History of power sources in endoscopic surgery. J Minim Invasive Gynecol 2013;20(3):271-8. 42. Venturella R, Morelli M, Lico D, Di Cello A, Rocca M, Sacchinelli A, et al. Wide
447
excision of soft tissues adjacent to the ovary and fallopian tube does not impair the
448
ovarian reserve in women undergoing prophylactic bilateral salpingectomy: results from
449
a randomized, controlled trial. Fertil Steril 2015;104(5):1332-9.
AC C
EP
TE D
M AN U
SC
RI PT
446
ACCEPTED MANUSCRIPT Pereira 24 Figures
451
Figure 1: Selection of study cohort.
452
Figure 2, Panel A: Comparison of ovarian stimulation duration (days) in different
453
salpingectomy groups and controls; Panel B: Comparison of total gonadotropins administered
454
(IU) in different salpingectomy groups and controls. *P value calculated from Analysis of
455
Variance test
456
Figure 3, Panel A: Comparison of total oocytes retrieved in different salpingectomy groups and
457
controls; Panel B: Comparison of fertilization rates in the subsequent in vitro fertilization cycle;
458
Panel C: Comparison of implantation and clinical pregnancy rates in different salpingectomy
459
groups and controls. *P value calculated from Analysis of Variance test
460
**P value calculated from Chi-Square test
AC C
EP
TE D
M AN U
SC
RI PT
450
ACCEPTED MANUSCRIPT Pereira 25 Acknowledgement
462
The authors would like to thank Dr. Paul J. Christos for his assistance with statistical analyses
463
and Logan Stone for his editorial assistance.
AC C
EP
TE D
M AN U
SC
RI PT
461
ACCEPTED MANUSCRIPT
Tables Table 1: Surgical characteristics of patients undergoing salpingectomy (n=145) Unilateral Hydrosalpinx (n=75)
16 (55.2%) 13 (44.8%)
44 (58.7%) 31 (41.3%)
Instrument Harmonic® LigaSureTM
10 (34.5%) 19 (65.5%)
30 (40%) 45 (60%)
Time between salpingectomy and ovarian stimulation (days)
54.2 (±12.9)
51.7 (±10.8)
6 (60%) 4 (40%)
4 (40%) 6 (60%)
12 (38.7%) 19 (61.3%)
55.5 (±9.49)
EP
TE D
Data are presented as mean ± standard deviation and n (%) *P value calculated from Chi-Square test **P value calculated from Analysis of Variance test
AC C
Other (n=10)
18 (58.1%) 13 (41.9%)
M AN U
Fallopian tube removed Right Left
Bilateral Hydrosalpinx (n=31)
RI PT
Ectopic Pregnancy (n=29)
SC
Parameter
P
0.92*
0.83*
0.12**
59.2 (±9.91)
ACCEPTED MANUSCRIPT
Table 2: Comparison of demographics and baseline IVF characteristics of patients undergoing ovarian stimulation after unilateral salpingectomy (n=145) with controls (n=125) Hydrosalpinx (n=75)
Bilateral Hydrosalpinx (n=31)
Age (years)
36.1 (±2.08)
35.9 (±1.97)
36.3 (±2.14)
Body Mass Index (kg/m2)
22.6 (±5.96)
22.7 (±5.75)
22.9 (±5.45)
Other (n=10)
Control (n=125)
P*
36.2 (±2.34)
35.4 (±2.21)
0.14
22.8 (±6.12)
22.1 (±5.37)
0.92
0.12
RI PT
Ectopic Pregnancy (n=29)
M AN U
SC
Parameter
Before Laparoscopic Salpingectomy and Ovarian Stimulation
AMH (ng/mL)
11.9 (±1.59)
11.1 (±1.89)
1.62 (±0.93)
1.70 (±0.71)
TE D
Antral Follicle Count
11.6 (±2.01)
11.4 (±2.09)
11.8 (±1.92)
1.79 (±0.43)
1.65 (±0.58)
1.73 (±0.79)
0.91
11.2 (±2.01)
Cycle Day 2/3 FSH (mIU/mL)
4.21 (±2.61)
Cycle Day 2/3 E2 (pg/mL)
51.5 (±19.1)
11.7 (±1.74)
11.4 (±1.79)
11.5 (±1.63)
11.3 (±1.81)
0.59
4.42 (±2.51)
4.31 (±1.99)
4.48 (±2.47)
4.19 (±2.84)
0.98
49.7 (±20.8)
54.2 (±18.2)
53.7 (±21.0)
50.2 (±17.6)
0.80
AC C
Antral Follicle Count
EP
After Laparoscopic Salpingectomy and Before Ovarian Stimulation
Data are presented as mean ± standard deviation; FSH: Follicle Stimulating Hormone; E2: Estradiol; AMH: anti-müllerian hormone *P value calculated from Analysis of Variance test
ACCEPTED MANUSCRIPT
Table 3: Multiple linear regression model to examine the relationship between salpingectomy indication and ovarian stimulation duration 95% Confidence Intervals
P
0.12
Robust Standard Error 0.04
Ectopic pregnancy vs. other salpingectomy indications (referent) Age (years)
0.007-0.015
<0.001
-0.25
0.17
-0.71, 0.19
0.19
BMI (kg/m2)
-0.002
0.006
-0.02, 0.01
0.78
Time elapsed between salpingectomy and ovarian stimulation (days)
0.0002
0.0001
-0.0002, 0.0002
0.85
AC C
EP
TE D
M AN U
BMI: Body mass index
RI PT
Coefficient
SC
Ovarian Stimulation Duration (days)
ACCEPTED MANUSCRIPT
Table 4: Multiple linear regression model to examine the relationship between salpingectomy indication and total gonadotropins administered. 95% Confidence Intervals
P
211.4
Robust Standard Error 36.0
Ectopic pregnancy vs. other salpingectomy indications (referent) Age (years)
118.8, 303.9
0.002
105.9
59.9
-46.2, 18.3
0.14
BMI (kg/m2)
-13.9
12.6
-1029.3, 399.9
0.32
Time elapsed between salpingectomy and ovarian stimulation (days)
-152.8
49.9
-281.3, 24.3
0.56
AC C
EP
TE D
M AN U
BMI: Body mass index
RI PT
Coefficient
SC
Total gonadotropins administered (IU)
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S1553-4650(17)30002-X
DOI:
10.1016/j.jmig.2016.12.023
Reference:
JMIG 3028
To appear in:
The Journal of Minimally Invasive Gynecology
Received Date: 30 September 2016 Revised Date:
24 December 2016
Accepted Date: 31 December 2016
Please cite this article as: Pereira N, Pryor KP, Voskuilen-Gonzalez A, Lekovich JP, Elias RT, Spandorfer SD, Rosenwaks Z, Ovarian response and in vitro fertilization outcomes after salpingectomy: does salpingectomy indication matter?, The Journal of Minimally Invasive Gynecology (2017), doi: 10.1016/j.jmig.2016.12.023. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Pereira 1 Ovarian response and in vitro fertilization outcomes after salpingectomy: does
2
salpingectomy indication matter?
3
Nigel Pereira, M.D.,1 Katherine P. Pryor, M.D., M.S.,2 Anna Voskuilen-Gonzalez, M.D.,1 Jovana
4
P. Lekovich, M.D.,1 Rony T. Elias, M.D.,1 Steven D. Spandorfer, M.D.,1 Zev Rosenwaks, M.D.1
5
1
6
Medicine, New York, New York.
7
2
8
York
SC
The Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell
M AN U
Department of Obstetrics and Gynecology, Weill Cornell Medical College, New York, New
9 10
RI PT
1
Financial Disclosures: None for all authors
11
Corresponding author: Nigel Pereira, M.D., The Ronald O. Perelman and Claudia Cohen Center
13
for Reproductive Medicine, Weill Cornell Medicine, 1305 York Ave., New York, New York
14
10021 (E-mail: [email protected]).
EP
16
Short title: Ovarian response and IVF outcomes stratified by salpingectomy indication
AC C
15
TE D
12
ACCEPTED MANUSCRIPT Pereira 2 Précis
18
Patients undergoing in vitro fertilization after salpingectomy for ectopic pregnancy, unilateral
19
hydrosalpinx, bilateral hydrosalpinges, or other indications have similar oocyte yield,
20
fertilization, implantation and clinical pregnancy rates.
AC C
EP
TE D
M AN U
SC
RI PT
17
ACCEPTED MANUSCRIPT Pereira 3 Abstract
22
Study Objective: To investigate whether the ovarian response and pregnancy outcomes of
23
patients undergoing in vitro fertilization (IVF) after salpingectomy is affected by the underlying
24
indication for salpingectomy.
25
Design: Retrospective cohort study.
26
Design Classification: II-3.
27
Setting: University-affiliated fertility center.
28
Patients: All patients <37 years of age undergoing IVF within 12 months of laparoscopic
29
salpingectomy. The underlying indication for laparoscopic salpingectomy in the study cohort
30
was either tubal ectopic pregnancy, unilateral or bilateral hydrosalpinx, or other (hematosalpinx
31
or pyosalpinx), as confirmed by histopathology.
32
Interventions: IVF and embryo transfer (ET).
33
Measurements and Main Results: Surgical characteristics, demographics, ovarian stimulation
34
parameters, total oocytes retrieved, fertilization rates, implantation rates, and clinical pregnancy
35
rates were compared among the salpingectomy groups. Age and time-matched patients
36
undergoing their first IVF-ET cycle for male factor infertility, with no prior history of
37
laparoscopy served as controls.
38
Results: Of the 996 laparoscopic procedures during the study period, 136 patients underwent
39
unilateral salpingectomy for the following indications – ectopic pregnancy: 39; unilateral
40
hydrosalpinx: 81; other: 16. Of these, 29 patients in the ectopic pregnancy group, 75 patients in
41
the unilateral hydrosalpinx group, and 10 patients in the ‘other’ group underwent subsequent
AC C
EP
TE D
M AN U
SC
RI PT
21
ACCEPTED MANUSCRIPT Pereira 4 IVF-ET. Thirty-one patients underwent both bilateral salpingectomy and subsequent IVF-ET.
43
There was no difference in the antral follicle counts before and after salpingectomy in all groups.
44
There was a statistical difference in the mean duration of ovarian stimulation in the
45
salpingectomy groups – ectopic pregnancy: 10.9 ±2.15 days; unilateral hydrosalpinx: 9.56 ±1.95
46
days; bilateral hydrosalpinx: 9.51 ±2.01 days; other: 9.89 ±2.20 days; control: 9.76 ±1.99 days.
47
Similar trends were noted for total gonadotropins administered when comparing the ectopic
48
pregnancy group (3375.9 ±931.0 IU) to the remaining groups (unilateral hydrosalpinx: 2841.3
49
±1160.9 IU; bilateral hydrosalpinx: 2519.3 ±1004.7 IU; other: 2808.6 ±990.1 IU; control: 2726.1
50
±1129.8 IU). There was no difference in the total number of oocytes retrieved, fertilization rate,
51
implantation rate or clinical pregnancy rate in the salpingectomy groups compared to controls.
52
Conclusions: Although our findings indicate that patients undergoing IVF after salpingectomy
53
for an ectopic pregnancy have a statistically longer duration of stimulation and require higher
54
gonadotropin doses than patients undergoing IVF after salpingectomy for other indications, these
55
differences are of limited clinical significance given that the total number of oocytes retrieved,
56
implantation rate, and clinical pregnancy rate among the different salpingectomy groups was
57
comparable to controls.
58
Key Words: salpingectomy; laparoscopy; reproductive surgery; ovarian reserve; ovarian
59
response; ectopic pregnancy; hydrosalpinx; in vitro fertilization
AC C
EP
TE D
M AN U
SC
RI PT
42
ACCEPTED MANUSCRIPT Pereira 5 60 61
Introduction Salpingectomy is a commonly performed gynecologic procedure that involves the surgical removal of one or both fallopian tubes. In reproductive-aged women, salpingectomy is
63
routinely performed for the management of tubal pathologies, especially ectopic pregnancy and
64
hydrosalpinx. Lately, it has also been utilized prophylactically in women undergoing
65
hysterectomy to reduce the risk of ovarian cancer (1-4). Recent studies have suggested that
66
salpingectomy may have deleterious effects on ovarian reserve in all women, as well as ovarian
67
responsiveness in women undergoing various fertility treatments (5-8). The mechanism
68
underlying these outcomes is the theoretical disruption of collateral blood supply to the ovary
69
(5,7). However, other studies have refuted any detrimental impact of salpingectomy on ovarian
70
reserve and response (9-14).
SC
M AN U
These contrasting results in the medical literature (5-14) can be attributed to marked
TE D
71
RI PT
62
variations in study methodology (retrospective vs. randomized controlled trials vs. meta-
73
analysis), sample sizes, patient population (reproductive-aged women vs. women undergoing
74
hysterectomy), markers of ovarian reserve (antral follicle counts vs. follicle stimulating hormone
75
levels vs. anti-müllerian hormone levels), and markers of ovarian responsiveness (oocytes
76
retrieved vs. gonadotropins administered). Furthermore, several studies investigating the effects
77
of salpingectomy on ovarian reserve and response have included other concomitant surgical
78
procedures or diagnoses that may confound results. Thus, the objective of the current study is to
79
investigate whether the ovarian response and pregnancy outcomes of patients undergoing in vitro
80
fertilization (IVF) after salpingectomy is affected by the underlying indication for
81
salpingectomy.
AC C
EP
72
ACCEPTED MANUSCRIPT Pereira 6 82
Materials and Methods
83
Inclusion and Exclusion Criteria All patients <37 years of age undergoing ovarian stimulation for in vitro fertilization
RI PT
84
(IVF) after laparoscopic surgery at our center between January 2008 and June 2015 were
86
assessed for potential inclusion. Of this initial cohort of patients, only those undergoing IVF
87
within 12 months of laparoscopic salpingectomy were included. By study design, patients
88
undergoing salpingectomy via a non-laparoscopic approach were excluded. In addition, patients
89
undergoing concomitant procedures such as hysterectomy or oophorectomy were also excluded.
90
All patients undergoing IVF had normal baseline ovarian reserve as highlighted by the following
91
criteria: cycle day-2/3 follicle stimulating hormone (FSH) level <12 mIU/mL, cycle day-2/3 anti-
92
müllerian hormone (AMH) level >1 ng/mL, and an antral follicle count (AFC) of 8-15 (15). The
93
institutional review board at Weill Cornell Medical College approved our study protocol.
94
Surgical Methods
M AN U
TE D
95
SC
85
The underlying indication for laparoscopic salpingectomy in the study cohort was either tubal ectopic pregnancy, unilateral or bilateral hydrosalpinx and ‘other’. An ectopic pregnancy
97
was diagnosed with standard transvaginal ultrasonographic criteria in the setting of abnormally
98
rising beta human chorionic gonadotropin (βhCG) levels (16,17). In general, patients with an
99
ectopic pregnancy >3.5 cm, βhCG level >5000 mIU/mL, fetal cardiac activity or abdominal pain
AC C
EP
96
100
underwent unilateral laparoscopic salpingectomy with histopathologic confirmation (18). All
101
patients with evidence of hydrosalpinx on transvaginal ultrasonography or
102
hysterosalpingography underwent unilateral or bilateral laparoscopic salpingectomy prior to IVF
103
(19,20). Patients were considered to have a hydrosalpinx if the ultrasonographic and
ACCEPTED MANUSCRIPT Pereira 7 104
histopathologic diagnoses were consistent. Patients with histopathologic evidence of
105
hematosalpinx or pyosalpinx after laparoscopic salpingectomy were considered to have ‘other.’ All salpingectomies were performed by 1 of 4 surgeons using a standard laparoscopic
RI PT
106
technique (18). Briefly, salpingectomies were performed by cauterizing the mesosalpinx along
108
the entire length of the fallopian tube, using either a 5-mm Harmonic® scalpel (Ethicon Endo-
109
Surgery, J&J Medical Ltd, Cincinnati, USA) or LigaSureTM device (Covidien, Dublin, Ireland).
110
Cauterization was performed beginning at the fimbriated end running to the cornual end, with
111
care taken to cauterize as closely as possible to the fallopian tube to minimize damage to the
112
tubal vascular supply. The proximal end of the fallopian tube was transected. The transected
113
fallopian tube was removed through a 12-mm port using an endoscopic bag.
114
Clinical Protocols
M AN U
Ovarian stimulation was performed with gonadotropins (Follistim, Merck, Kenilworth,
TE D
115
SC
107
NJ, USA; Gonal-F, EMD-Serono Inc., Rockland, MA, USA; and Menopur, Ferring
117
Pharmaceuticals Inc, Parsippany, NJ, USA) to maximize follicular response while minimizing
118
the risk of ovarian hyperstimulation syndrome (21). Gonadotropin doses were based on patient
119
age, weight, antral follicle count (AFC), and serum anti-müllerian hormone (AMH) levels.
120
Ovulation was suppressed using daily injections of 0.25 mg Ganirelix Acetate (Merck,
121
Kenilworth, NJ, USA). hCG was used as the ovulatory trigger, which was given when the two
122
lead follicles attained a mean diameter >17 mm. Oocyte retrieval was performed under conscious
123
sedation using a 30 cm 16 G oocyte aspiration needle (Cook® Medical, Bloomington, IN, USA)
124
using transvaginal ultrasound guidance approximately 35–37 hours after hCG administration.
AC C
EP
116
ACCEPTED MANUSCRIPT Pereira 8 125
The retrieved oocytes were fertilized using conventional in vitro insemination or intracytoplasmic sperm injection (ICSI) depending on the couple’s reproductive history. Oocytes
127
were examined 14-18 hours after conventional insemination or ICSI to assess the fertilization
128
rate i.e., number of 2-pronuclear embryos out of the number of mature oocytes inseminated or
129
injected (22). The resulting embryos were cultured in in-house media and then graded at the
130
cleavage stage using the Veeck criteria (23). All fresh ETs were performed on day 3 (cleavage-
131
stage) with Wallace catheters (Smiths Medical Inc., Norwell, MA, USA) at approximately 1 cm
132
less than the uterine depth identified at prior trial transfer (24). Ultrasound guidance was utilized
133
only when ETs were deemed difficult based on the prior trial transfer (24). Surplus embryos
134
were cultured to the blastocyst-stage and the top-quality blastocysts were cryopreserved by
135
vitrification (25). No patients underwent pre-implantation genetic screening of embryos in the
136
current study.
137
Variables Recorded
SC
M AN U
TE D
138
RI PT
126
Surgical characteristics recorded were the laterality of the excised fallopian tube (right vs. left), surgeon performing the laparoscopic salpingectomy, the laparoscopic instrument utilized
140
(Harmonic® scalpel or LigaSureTM device) and the final histopathologic diagnosis. The time
141
between the laparoscopic salpingectomy and initiation of ovarian stimulation was also recorded.
142
Baseline demographic characteristics recorded for all patients included age and body mass index
143
(BMI, kg/m2). Cycle day 2/3 AFC and AMH levels (ng/mL) were measured in all patients prior
144
to laparoscopy. The baseline IVF parameters of all patients prior to ovarian stimulation were also
145
noted, which included AFC, cycle day 2/3 follicle stimulating hormone (FSH, mIU/mL) level,
146
cycle day 2/3 estradiol (E2, pg/mL) level, and infertility diagnosis. Ovarian stimulation
147
parameters assessed were the duration of ovarian stimulation (days), total gonadotropins
AC C
EP
139
ACCEPTED MANUSCRIPT Pereira 9 administered (IU), and total oocytes retrieved. Finally, the following IVF outcomes were also
149
noted: fertilization rate (%), mean number of embryos transferred, implantation rate defined as
150
the mean number of sacs seen on ultrasonography divided by the number of embryos transferred
151
for each patient, and clinical pregnancy rate defined as the number of intrauterine gestations with
152
fetal cardiac activity per IVF cycle. The number of surplus embryos cryopreserved at the
153
blastocyst-stage was recorded as well.
154
Statistical Analyses
SC
Continuous variables were assessed for normality by the Shapiro-Wilk’s test and all were
M AN U
155
RI PT
148
found to be normally distributed; all continuous variables are expressed as mean ± standard
157
deviation (SD). Categorical variables were expressed as number of cases (n) and percentage of
158
occurrence (%). Patients undergoing salpingectomy were stratified into 4 groups based on
159
histopathologic diagnosis i.e., ectopic pregnancy, unilateral hydrosalpinx, bilateral hydrosalpinx
160
and other. Age and time-matched patients undergoing their first IVF-ET cycle for male factor
161
infertility, with no prior history of laparoscopy served as controls. Baseline demographics,
162
baseline IVF parameters and the time between laparoscopic salpingectomy and initiation of
163
ovarian stimulation were compared between groupings of histopathologic diagnosis and controls
164
by using analysis of variance (ANOVA) tests. Chi-square (χ2) tests were used to compare
165
fertilization, implantation and clinical pregnancy rates between groupings of histopathologic
166
diagnosis and controls. Multiple linear regression models were used to investigate the
167
relationship between 1) salpingectomy indication and ovarian stimulation duration and 2)
168
salpingectomy indication and total gonadotropins administered, controlling for age, BMI, and
169
time elapsed between salpingectomy indication and ovarian stimulation. Also, in order to address
170
the potential issue of surgeon variability in performing the laparoscopic salpingectomy (i.e.,
AC C
EP
TE D
156
ACCEPTED MANUSCRIPT Pereira 10 potential clustering of outcomes within surgeon), robust standard errors were computed by using
172
generalized estimating equations (GEE) modeling, specifically clustering on performing surgeon.
173
In a previous investigation, we reported a gonadotropin dosage difference of 705 IU in patients
174
undergoing IVF after unilateral salpingectomy i.e., 2124 IU pre-salpingectomy and 3126 IU
175
post-salpingectomy (18). Based on these findings, a sample size of 27 patients per group was
176
estimated assuming an alpha level of 5% and a power of 80%. All statistical analyses were
177
performed using STATA version 14 (StataCorp, College Station, TX). Statistical significance
178
was set at P<0.05.
179
Results
M AN U
SC
RI PT
171
Nine hundred ninety-six patients underwent laparoscopy during the study period: 136
181
(13.7%) unilateral salpingectomy, 31 (3.11%) bilateral salpingectomy, and 829 (83.2%) other
182
laparoscopic procedures. Among patients undergoing unilateral salpingectomy, 39 (28.7%)
183
patients had an ectopic pregnancy, 81 (59.6%) patients had a hydrosalpinx, while the remaining
184
16 (11.7%) had other histopathologic diagnoses. Twenty-nine out of the 39 patients in the
185
ectopic pregnancy group, 75 out of the 81 patients in the hydrosalpinx group, and 10 out of the
186
16 patients in the ‘other’ group underwent subsequent ovarian stimulation. Of note, the final
187
histopathologic diagnosis in the other group was hematosalpinx (8 patients) and pyosalpinx (1
188
patient). The tenth patient underwent unilateral salpingectomy for spontaneous torsion of the
189
fallopian tube. All 31 patients in the bilateral salpingectomy group underwent salpingectomy and
190
subsequent ovarian stimulation. Figure 1 summarizes the selection of the study cohort.
191 192
AC C
EP
TE D
180
Table 1 compares the surgical characteristics of all patients undergoing salpingectomy stratified by salpingectomy indication. No difference was noted in the laterality of fallopian tube
ACCEPTED MANUSCRIPT Pereira 11 removed (right vs. left), type of surgical instrument used, or the time elapsed between
194
salpingectomy and subsequent ovarian stimulation. Of note, patients in the ectopic pregnancy,
195
unilateral hydrosalpinx and ‘other’ groups had underlying male factor or unexplained infertility;
196
patients in the bilateral hydrosalpinx group were considered to have tubal infertility. As evident
197
from table 2, no differences were noted in the baseline demographics of patients undergoing
198
salpingectomy. Furthermore, the AFC and AMH levels of all salpingectomy groups were similar,
199
indicating that the groups were comparable for ovarian reserve at baseline. Table 2 also shows
200
that the AFC, FSH and E2 levels of the salpingectomy groups were similar to controls even after
201
laparoscopy, suggesting no acute changes in ovarian reserve markers.
M AN U
SC
RI PT
193
Figure 2 (panel A) compares the duration of ovarian stimulation in patients after
203
salpingectomy. There was a statistically significant difference in the mean duration of ovarian
204
stimulation in the ectopic pregnancy group (10.9 ±2.15 days) compared to the unilateral
205
hydrosalpinx (9.56 ±1.95 days), bilateral hydrosalpinx (9.51 ±2.01 days), other (9.89 ±2.20
206
days), and control (9.76 ±1.99 days) groups (P=0.01). As seen in figure 2 (panel B), similar
207
trends were noted for the total gonadotropins administered when comparing the ectopic
208
pregnancy group (3375.9 ±931.0 IU) to the remaining groups (unilateral hydrosalpinx: 2841.3
209
±1160.9 IU; bilateral hydrosalpinx: 2519.3 ±1004.7 IU; other: 2808.6 ±990.1 IU; control: 2726.1
210
±1129.8 IU; P<0.001). The duration of ovarian stimulation and dosage of gonadotropins was
211
+0.12 days longer (table 3) and +211.4 IU higher (table 4), respectively in the ectopic group
212
compared to all other salpingectomy groups, after multiple linear regression accounting for age,
213
BMI and the time elapsed between salpingectomy and IVF-ET. Figure 3 (panel A) shows no
214
difference in the total number of oocytes retrieved in the salpingectomy groups compared to
215
controls. Figure 3 (panel B) indicates that there were no differences in the fertilization rate.
AC C
EP
TE D
202
ACCEPTED MANUSCRIPT Pereira 12 Furthermore, Figure 3 (panel C) also reveals that the mean embryos transferred, mean
217
implantation rates and clinical pregnancy rates were comparable between the salpingectomy
218
groups. The mean number of top-quality embryos cryopreserved at the blastocyst-stage in the
219
different salpingectomy groups was comparable and as follows: 1.72 ±0.99 (ectopic pregnancy);
220
1.94 ±0.88 (unilateral salpingectomy); 1.95 ±1.03 (bilateral salpingectomy); 1.87 ±0.98 (other);
221
1.80 ±0.87 (control).
222
Discussion
SC
RI PT
216
The current retrospective study investigated whether ovarian response or pregnancy
224
outcomes in patients undergoing IVF-ET after salpingectomy is affected by the underlying
225
indication for salpingectomy. While our findings suggest that patients undergoing IVF after
226
salpingectomy for an ectopic pregnancy have a statistically longer duration of ovarian
227
stimulation and require statistically higher doses of gonadotropins than patients undergoing IVF
228
after salpingectomy for other indications, these differences are of limited clinical significance
229
given that the total number of oocytes retrieved, fertilization rate, implantation rate, and clinical
230
pregnancy rate among the different salpingectomy groups was similar to controls.
TE D
EP
231
M AN U
223
Whether or not salpingectomy impairs ovarian reserve remains controversial. Several conflicting studies have investigated the effect of salpingectomy on ovarian reserve by looking at
233
surrogate markers in subsequent ovarian stimulation cycles. In a prospective study of 29 patients
234
who had undergone unilateral salpingectomy for management of ectopic pregnancy, Lass et al.
235
(26) noted decreased follicular development and fewer oocytes retrieved from the ovary on the
236
operated side. Similarly, Gelbaya et al. (27) noted significantly increased baseline FSH levels,
237
decreased E2 levels at time of trigger, decreased follicular development and fewer oocytes
AC C
232
ACCEPTED MANUSCRIPT Pereira 13 retrieved following salpingectomy performed for hydrosalpinx. Xi et al. (12) showed that women
239
undergoing salpingectomy for ectopic pregnancy required higher initial and total doses of
240
gonadotropins during subsequent IVF cycles as compared to non-operated controls. In contrast,
241
Ye at al. (5) did not observe differences in gonadotropin dosing, but instead noted significantly
242
decreased AMH in women after bilateral salpingectomy. Finally, Chan et al. (28) noted
243
decreased AFC and ovarian stromal blood flow on the operated side. At first glance, the
244
aforementioned studies may suggest that salpingectomy impairs ovarian reserve; however, there
245
are several other prospective and retrospective studies that have found no differences in ovarian
246
reserve following unilateral or bilateral salpingectomy performed for the management of either
247
ectopic pregnancy or hydrosalpinx (9-14, 29-32). Supplemental table 1 summarizes all studies
248
evaluating ovarian reserve or responsiveness after salpingectomy for ectopic pregnancy or
249
hydrosalpinx. As evident from the table, the large majority of studies have shown no difference
250
in ovarian reserve or responsiveness after salpingectomy for either indication.
SC
M AN U
TE D
251
RI PT
238
In a recent meta-analysis of 25 studies that included 1935 patients, Fan and Ma (7) suggested that salpingectomy impaired ovarian function. Specifically, when comparing subjects
253
undergoing unilateral salpingectomy to controls, there was a statistically significant decrease in
254
the number of oocytes retrieved. When comparing subjects undergoing bilateral salpingectomy
255
to controls, longer gonadotropin stimulation and higher doses were required, baseline FSH was
256
increased, and the number of oocytes retrieved was decreased. No differences were observed
257
when comparing the unilateral and bilateral salpingectomy groups. As the authors noted, the
258
meta-analysis was complicated by significant heterogeneity of the studies, including variability
259
in salpingectomy technique, as well as the time elapsed between surgery and subsequent ovarian
AC C
EP
252
ACCEPTED MANUSCRIPT Pereira 14 260
stimulation. Notably, the meta-analysis did not stratify results by the indication for
261
salpingectomy. Salpingectomy has been postulated to decrease ovarian reserve through several
RI PT
262
mechanisms. One common argument is that ovarian blood flow can be compromised during
264
salpingectomy. Because the tubal branch of the uterine artery originates at the same point as the
265
ovarian branch of the uterine artery, damage to ovarian branch as it traverses the mesosalpinx, by
266
surgical disruption or thermal spread can decrease blood flow to the ovary (18,27). It is also
267
possible that certain changes in blood flow to the fallopian tubes and ovaries through the
268
mesosalpinx may be unique to ectopic pregnancies. In ectopic pregnancies, trophoblastic
269
invasion of the fallopian tube increases blood flow through the tubal branch of the uterine artery
270
and promotes development of collateral blood flow (35). In addition, tubal pregnancies have
271
been shown to affect the pulsatility of the ipsilateral uterine artery (36). Thus, trophoblastic
272
invasion of the fallopian tube can affect ovarian blood flow such that at the time of
273
salpingectomy, the ovary has increased vulnerability to abrupt changes in blood supply. In
274
contrast, hemodynamic studies using Doppler ultrasonography in patients with hydrosalpinges
275
have revealed that the blood flow to the endometrium and ovary may be impaired at baseline
276
(37,38). Thus, an abrupt decrease in blood flow to ovary may not occur after salpingectomy for
277
hydrosalpinges as compared to tubal ectopic pregnancies.
M AN U
TE D
EP
AC C
278
SC
263
Some surgeons have also argued that disruption of blood flow may occur only with poor
279
surgical technique (39). Indeed, surgical technique and the type of laparoscopic instrument used
280
can also mediate the association between salpingectomy and reduced ovarian responsiveness
281
(39). However, as previously described, we used either a 5-mm Harmonic® scalpel or
282
LigaSureTM device to cauterize the mesosalpinx as close as possible to the fallopian tube to
ACCEPTED MANUSCRIPT Pereira 15 minimize inadvertent damage to ovarian blood supply. Furthermore, these instruments minimize
284
lateral thermal spread and instrument heating compared to monopolar or conventional bipolar
285
electrocautery, thereby reducing unintended tissue damage (40, 41).
286
RI PT
283
Although disruption in ovarian blood flow and surgical technique may contribute to the pathogenesis of decreased ovarian reserve after unilateral or bilateral salpingectomy, it is
288
important to highlight the existence of well-designed studies that rebut such results. One such
289
study is by Morelli et al. (14), in which the authors compared blood ovarian reserve parameters
290
(AMH and FSH levels) and ultrasonographic ovarian reserve parameters (AFC, mean ovarian
291
diameters and peak systolic velocity) in 79 patients 3 months before and after total laparoscopic
292
hysterectomy, with or without salpingectomy. Patients who underwent total laparoscopic
293
hysterectomy with salpingectomy showed no differences in blood or ultrasonographic ovarian
294
reserve parameters compared to those who underwent laparoscopic hysterectomy without
295
salpingectomy. Furthermore, both groups had comparable operative times, blood loss,
296
postoperative hospital stay and postoperative return to normal activity. A subsequent randomized
297
controlled trial by Venturella et al. (42) compared 91 patients undergoing standard laparoscopic
298
salpingectomy to 95 patients undergoing laparoscopic salpingectomy with wide excision of the
299
mesosalpinx. The authors reported no differences in AMH level, FSH levels, AFC, and vascular,
300
flow and vascular-flow indices. Similar to the Morelli et al. study, no differences in peri-
301
operative or post-operative outcomes were noted.
M AN U
TE D
EP
AC C
302
SC
287
It is important to emphasize that an association, if any, between impaired ovarian
303
response after salpingectomy is multi-factorial, including publication biases. The main strength
304
of the current study is the demonstration of similar oocyte yield, fertilization rates, implantation
305
rates and clinical pregnancy rates in patients undergoing IVF-ET after laparoscopic
ACCEPTED MANUSCRIPT Pereira 16 salpingectomy, irrespective of the underlying indication for salpingectomy. In addition, the
307
current study also accounted for several surgical variables (laterality of the fallopian tube,
308
surgical instrumentation, surgeon performing the laparoscopy) and the time elapsed between
309
laparoscopic salpingectomy and IVF, which have not been evaluated by prior studies.
310
RI PT
306
Despite these strengths, our study was limited by its retrospective nature and the small sample sizes of the salpingectomy groups. While the sample sizes were estimated to find
312
differences in gonadotropin dosage, post-hoc calculations suggest an approximate sample of size
313
of 376 per salpingectomy group to detect a 10% difference in clinical pregnancy rates. It is
314
worthwhile to note the unusually small magnitude of variation in the means and standard
315
deviations of all parameters listed in table 2. While this lack of variation can be attributed to the
316
normal baseline ovarian reserve parameters of all patients, the small sample size may also be an
317
overall source for such lack of variation. Another shortcoming of our study is that post-
318
salpingectomy AMH levels were not measured; however, it is reassuring to note that post-
319
salpingectomy AFC, FSH and E2 levels were similar to controls. Finally, ovarian response was
320
not evaluated separately in the ipsilateral ovary and the contralateral ovary when unilateral
321
salpingectomies were performed. While we acknowledge that study’s overall generalizability is
322
limited due to the aforementioned limitations, we deem that these findings do merit further
323
investigation in a prospective setting.
M AN U
TE D
EP
AC C
324
SC
311
In conclusion, the current study demonstrates similar oocyte yield and pregnancy
325
outcomes in patients undergoing IVF after salpingectomy, irrespective of the indication for
326
salpingectomy. Although the overall sample size remains small, the study did account for several
327
surgical variables such as laterality of the fallopian tube, surgical instrument used, performing
328
surgeon as well as the time elapsed between salpingectomy and ovarian stimulation. Thus, our
ACCEPTED MANUSCRIPT Pereira 17 findings can be useful to reassure patients undergoing ovarian stimulation for IVF-ET after
330
unilateral or bilateral laparoscopic salpingectomy.
AC C
EP
TE D
M AN U
SC
RI PT
329
ACCEPTED MANUSCRIPT Pereira 18
333 334 335 336
1. American College of Obstetricians and Gynecologists. Salpingectomy for ovarian cancer prevention. Committee Opinion No. 620. Obstet Gynecol 2015;125:279–81.
RI PT
332
References
2. Falconer H, Yin L, Grönberg H, Altman D. Ovarian cancer risk after salpingectomy: a nationwide population-based study. J Natl Cancer Inst 2015;107(2).
3. McAlpine JN, Hanley GE, Woo MM, Tone AA, Rozenberg N, Swenerton KD et al.
SC
331
Opportunistic salpingectomy: uptake, risks, and complications of a regional initiative for
338
ovarian cancer prevention. Am J Obstet Gynecol 2014;210(5):471.e1-11.
339
M AN U
337
4. Yoon SH, Kim SN, Shim SH, Kang SB, Lee SJ. Bilateral salpingectomy can reduce the
340
risk of ovarian cancer in the general population: A meta-analysis. Eur J Cancer
341
2016;55:38-46.
342
5. Ye XP, Yang YZ, Sun XX. A retrospective analysis of the effect of salpingectomy on serum antiMüllerian hormone level and ovarian reserve. Am J Obstet Gynecol
344
2015;212(1):53.e1-10.
345
TE D
343
6. Grynnerup AG, Lindhard A, Sørensen S. Anti-Müllerian hormone levels in salpingectomized compared with nonsalpingectomized women with tubal factor
347
infertility and women with unexplained infertility. Acta Obstet Gynecol Scand
349 350 351
AC C
348
EP
346
2013;92(11):1297-303.
7. Fan M, Ma L. Effect of salpingectomy on ovarian response to hyperstimulation during in vitro fertilization: a meta-analysis. Fertil Steril 2016;106(2):322-329.e9.
8. Orvieto R, Saar-Ryss B, Morgante G, Gemer O, Anteby EY, Meltcer S. Does
352
salpingectomy affect the ipsilateral ovarian response to gonadotropin during in vitro
353
fertilization-embryo transfer cycles? Fertil Steril 2011;95(5):1842-4.
ACCEPTED MANUSCRIPT Pereira 19 354 355 356
9. Qin F, Du DF, Li XL. The Effect of Salpingectomy on Ovarian Reserve and Ovarian Function. Obstet Gynecol Surv 2016;71(6):369-76. 10. Yoon SH, Lee JY, Kim SN, Chung HW, Park SY, Lee C. Does salpingectomy have a deleterious impact on ovarian response in in vitro fertilization cycles? Fertil Steril 2016
358
Jun 20. pii: S0015-0282(16)61300-6.
RI PT
357
11. Lin YJ, Ou YC, Huang FJ, Lin PY, Kung FT, Lan KC. Ovarian response to
360
gonadotropins in patients with tubal factor infertility: salpingectomy versus
361
nonsalpingectomy. J Minim Invasive Gynecol 2013;20(5):637-41.
12. Xi W, Gong F, Tang Y, Zhang H, Lu G. Ovarian response to gonadotropins after
M AN U
362
SC
359
363
laparoscopic salpingectomy for ectopic pregnancy. Int J Gynaecol Obstet
364
2012;116(2):93-6.
13. Almog B, Wagman I, Bibi G, Raz Y, Azem F, Groutz A, et al. Effects of salpingectomy
366
on ovarian response in controlled ovarian hyperstimulation for in vitro fertilization: a
367
reappraisal. Fertil Steril 2011;95(8):2474-6.
TE D
365
14. Morelli M, Venturella R, Mocciaro R, Di Cello A, Rania E, Lico D, et al. Prophylactic
369
salpingectomy in premenopausal low-risk women for ovarian cancer: primum non
370
nocere. Gynecol Oncol 2013;129(3):448-51.
EP
368
15. Pereira N, Friedman C, Hutchinson AP, Lekovich JP, Elias RT, Rosenwaks Z. Increased
372
odds of live birth in fresh in vitro fertilization cycles with shorter ovarian stimulation.
373 374
375 376
AC C
371
Fertil Steril 2016; pii: S0015-0282(16)62875-3. doi: 10.1016/j.fertnstert.2016.09.044. [Epub ahead of print]
ACCEPTED MANUSCRIPT Pereira 20 377
16. Barnhart KT. Clinical practice. Ectopic pregnancy. N Engl J Med 2009;361(4):379-87.
378
17. Kirk E, Papageorghiou AT, Condous G, Tan L, Bora S, Bourne T. The diagnostic effectiveness of an initial transvaginal scan in detecting ectopic pregnancy. Hum Reprod
380
2007;22(11):2824-8.
RI PT
379
18. Pereira N, Gerber D, Gerber RS, Lekovich JP, Elias RT, Spandorfer SD, Rosenwaks Z.
382
Effect of Methotrexate or Salpingectomy for Ectopic Pregnancy on Subsequent In Vitro
383
Fertilization-Embryo Transfer Outcomes. J Minim Invasive Gynecol 2015;22(5):870-6.
SC
381
19. Patel MD, Acord DL, Young SW. Likelihood ratio of sonographic findings in
385
discriminating hydrosalpinx from other adnexal masses. AJR Am J Roentgenol
386
2006;186(4):1033-8.
387
M AN U
384
20. Johnson N, van Voorst S, Sowter MC, Strandell A, Mol BW. Surgical treatment for tubal disease in women due to undergo in vitro fertilization. Cochrane Database Syst Rev
389
2010;(1):CD002125.
392 393 394 395 396 397 398
2014;1154:171-231.
22. Palermo GD, Neri QV, Rosenwaks Z. To ICSI or Not to ICSI. Semin Reprod Med
EP
391
21. Huang JY, Rosenwaks Z. Assisted reproductive techniques. Methods Mol Biol
2015;33(2):92-102.
23. Gosden LV. Oocyte retrieval and quality evaluation. Methods Mol Biol 2014;1154:343-
AC C
390
TE D
388
60.
24. Pereira N, Petrini AC, Lekovich JP, Schattman GL, Rosenwaks Z. Singleton pregnancy and perinatal outcomes after fresh versus frozen-thawed blastocyst transfers. Int J Gynaecol Obstet 2016;135(1):96-100.
ACCEPTED MANUSCRIPT Pereira 21 399 400 401
25. Kuwayama M. Highly efficient vitrification for cryopreservation of human oocytes and embryos: the Cryotop method. Theriogenology 2007;67(1):73-80. 26. Lass A, Ellenbogen A, Croucher C, Trew G, Margara R, Becattini C, Winston RM. Effect of salpingectomy on ovarian response to superovulation in an in vitro fertilization-
403
embryo transfer program. Fertil Steril 1998;70(6):1035-8.
404
RI PT
402
27. Gelbaya TA, Nardo LG, Fitzgerald CT, Horne G, Brison DR, Lieberman BA. Ovarian response to gonadotropins after laparoscopic salpingectomy or the division of fallopian
406
tubes for hydrosalpinges. Fertil Steril 2006;85(5):1464-8.
28. Chan CC, Ng EH, Li CF, Ho PC. Impaired ovarian blood flow and reduced antral follicle
M AN U
407
SC
405
408
count following laparoscopic salpingectomy for ectopic pregnancy. Hum Reprod
409
2003;18(10):2175-80.
410
29. Strandell A, Lindhard A, Waldenstrom U, Thorburn J, Janson PO, Hamberger L. Hydrosalpinx and IVF outcome: a prospective, randomized multicentre trial in
412
Scandinavia on salpingectomy prior to IVF. Hum Reprod 1999;14(11):2762-9.
413
TE D
411
30. Dar P, Sachs GS, Strassburger D, Bukovsky I, Arieli S. Ovarian function before and after salpingectomy in artificial reproductive technology patients. Hum Reprod
415
2000;15(1):142-4.
EP
414
31. Demir B, Bozdag G, Sengul O, Kahyaoglu I, Mumusoglu S, Zengin D. The impact of
417
unilateral salpingectomy on antral follicle count and ovarian response in ICSI cycles:
418 419
AC C
416
comparison of contralateral side. Gynecol Endocrinol 2016:1-4.
32. Tal J, Paltieli Y, Korobotchka R, Ziskind G, Eibschitz I, Ohel G. Ovarian response to
420
gonadotropin stimulation in repeated IVF cycles after unilateral salpingectomy. J Assist
421
Reprod Genet 2002;19(10):451-5.
ACCEPTED MANUSCRIPT Pereira 22 422
33. Kontoravdis A, Makrakis E, Pantos K, Botsis D, Deligeoroglou E, Creatsas G. Proximal
423
tubal occlusion and salpingectomy result in similar improvement in in vitro fertilization
424
outcome in patients with hydrosalpinx. Fertil Steril 2006;86(6):1642-9. 34. Verhulst G, Vandersteen N, van Steirteghem AC, Devroey P. Bilateral salpingectomy
RI PT
425 426
does not compromise ovarian stimulation in an in-vitro fertilization/embryo transfer
427
programme. Hum Reprod 1994;9(4):624-8.
35. Kirchler HC, Seebacher S, Alge AA, Muller-Holzner E, Fessler S, Kolle D. Early
SC
428
diagnosis of tubal pregnancy: changes in tubal blood flow evaluated by endovaginal color
430
Doppler sonography. Obstet Gynecol 1993;82(4 Pt 1):561-5
431 432
M AN U
429
36. Tekay A, Jouppila P. Color Doppler flow as an indicator of trophoblastic activity in tubal pregnancies detected by transvaginal ultrasound. Obstet Gynecol 1992;80(6):995-9. 37. El-Mazny A, Ramadan W, Kamel A, Gad-Allah S. Effect of hydrosalpinx on uterine and
434
ovarian hemodynamics in women with tubal factor infertility. Eur J Obstet Gynecol
435
Reprod Biol 2016;199:55-9.
436
TE D
433
38. Ng EH, Chan CC, Tang OS, Ho PC. Comparison of endometrial and subendometrial blood flows among patients with and without hydrosalpinx shown on scanning during in
438
vitro fertilization treatment. Fertil Steril 2006;85(2):333-8.
440
39. Dietl J, Wischhusen J. The postreproductive salpingectomy. Fertil Steril
AC C
439
EP
437
2014;101(3):e20.
441
40. Sutton PA, Awad S, Perkins AC, Lobo DN. Comparison of lateral thermal spread using
442
monopolar and bipolar diathermy, the Harmonic Scalpel and the Ligasure. Br J Surg
443
2010;97(3):428-33.
ACCEPTED MANUSCRIPT Pereira 23 444 445
41. Sutton C, Abbott J. History of power sources in endoscopic surgery. J Minim Invasive Gynecol 2013;20(3):271-8. 42. Venturella R, Morelli M, Lico D, Di Cello A, Rocca M, Sacchinelli A, et al. Wide
447
excision of soft tissues adjacent to the ovary and fallopian tube does not impair the
448
ovarian reserve in women undergoing prophylactic bilateral salpingectomy: results from
449
a randomized, controlled trial. Fertil Steril 2015;104(5):1332-9.
AC C
EP
TE D
M AN U
SC
RI PT
446
ACCEPTED MANUSCRIPT Pereira 24 Figures
451
Figure 1: Selection of study cohort.
452
Figure 2, Panel A: Comparison of ovarian stimulation duration (days) in different
453
salpingectomy groups and controls; Panel B: Comparison of total gonadotropins administered
454
(IU) in different salpingectomy groups and controls. *P value calculated from Analysis of
455
Variance test
456
Figure 3, Panel A: Comparison of total oocytes retrieved in different salpingectomy groups and
457
controls; Panel B: Comparison of fertilization rates in the subsequent in vitro fertilization cycle;
458
Panel C: Comparison of implantation and clinical pregnancy rates in different salpingectomy
459
groups and controls. *P value calculated from Analysis of Variance test
460
**P value calculated from Chi-Square test
AC C
EP
TE D
M AN U
SC
RI PT
450
ACCEPTED MANUSCRIPT Pereira 25 Acknowledgement
462
The authors would like to thank Dr. Paul J. Christos for his assistance with statistical analyses
463
and Logan Stone for his editorial assistance.
AC C
EP
TE D
M AN U
SC
RI PT
461
ACCEPTED MANUSCRIPT
Tables Table 1: Surgical characteristics of patients undergoing salpingectomy (n=145) Unilateral Hydrosalpinx (n=75)
16 (55.2%) 13 (44.8%)
44 (58.7%) 31 (41.3%)
Instrument Harmonic® LigaSureTM
10 (34.5%) 19 (65.5%)
30 (40%) 45 (60%)
Time between salpingectomy and ovarian stimulation (days)
54.2 (±12.9)
51.7 (±10.8)
6 (60%) 4 (40%)
4 (40%) 6 (60%)
12 (38.7%) 19 (61.3%)
55.5 (±9.49)
EP
TE D
Data are presented as mean ± standard deviation and n (%) *P value calculated from Chi-Square test **P value calculated from Analysis of Variance test
AC C
Other (n=10)
18 (58.1%) 13 (41.9%)
M AN U
Fallopian tube removed Right Left
Bilateral Hydrosalpinx (n=31)
RI PT
Ectopic Pregnancy (n=29)
SC
Parameter
P
0.92*
0.83*
0.12**
59.2 (±9.91)
ACCEPTED MANUSCRIPT
Table 2: Comparison of demographics and baseline IVF characteristics of patients undergoing ovarian stimulation after unilateral salpingectomy (n=145) with controls (n=125) Hydrosalpinx (n=75)
Bilateral Hydrosalpinx (n=31)
Age (years)
36.1 (±2.08)
35.9 (±1.97)
36.3 (±2.14)
Body Mass Index (kg/m2)
22.6 (±5.96)
22.7 (±5.75)
22.9 (±5.45)
Other (n=10)
Control (n=125)
P*
36.2 (±2.34)
35.4 (±2.21)
0.14
22.8 (±6.12)
22.1 (±5.37)
0.92
0.12
RI PT
Ectopic Pregnancy (n=29)
M AN U
SC
Parameter
Before Laparoscopic Salpingectomy and Ovarian Stimulation
AMH (ng/mL)
11.9 (±1.59)
11.1 (±1.89)
1.62 (±0.93)
1.70 (±0.71)
TE D
Antral Follicle Count
11.6 (±2.01)
11.4 (±2.09)
11.8 (±1.92)
1.79 (±0.43)
1.65 (±0.58)
1.73 (±0.79)
0.91
11.2 (±2.01)
Cycle Day 2/3 FSH (mIU/mL)
4.21 (±2.61)
Cycle Day 2/3 E2 (pg/mL)
51.5 (±19.1)
11.7 (±1.74)
11.4 (±1.79)
11.5 (±1.63)
11.3 (±1.81)
0.59
4.42 (±2.51)
4.31 (±1.99)
4.48 (±2.47)
4.19 (±2.84)
0.98
49.7 (±20.8)
54.2 (±18.2)
53.7 (±21.0)
50.2 (±17.6)
0.80
AC C
Antral Follicle Count
EP
After Laparoscopic Salpingectomy and Before Ovarian Stimulation
Data are presented as mean ± standard deviation; FSH: Follicle Stimulating Hormone; E2: Estradiol; AMH: anti-müllerian hormone *P value calculated from Analysis of Variance test
ACCEPTED MANUSCRIPT
Table 3: Multiple linear regression model to examine the relationship between salpingectomy indication and ovarian stimulation duration 95% Confidence Intervals
P
0.12
Robust Standard Error 0.04
Ectopic pregnancy vs. other salpingectomy indications (referent) Age (years)
0.007-0.015
<0.001
-0.25
0.17
-0.71, 0.19
0.19
BMI (kg/m2)
-0.002
0.006
-0.02, 0.01
0.78
Time elapsed between salpingectomy and ovarian stimulation (days)
0.0002
0.0001
-0.0002, 0.0002
0.85
AC C
EP
TE D
M AN U
BMI: Body mass index
RI PT
Coefficient
SC
Ovarian Stimulation Duration (days)
ACCEPTED MANUSCRIPT
Table 4: Multiple linear regression model to examine the relationship between salpingectomy indication and total gonadotropins administered. 95% Confidence Intervals
P
211.4
Robust Standard Error 36.0
Ectopic pregnancy vs. other salpingectomy indications (referent) Age (years)
118.8, 303.9
0.002
105.9
59.9
-46.2, 18.3
0.14
BMI (kg/m2)
-13.9
12.6
-1029.3, 399.9
0.32
Time elapsed between salpingectomy and ovarian stimulation (days)
-152.8
49.9
-281.3, 24.3
0.56
AC C
EP
TE D
M AN U
BMI: Body mass index
RI PT
Coefficient
SC
Total gonadotropins administered (IU)
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT