Human Reproduction vol 11 no 9 pp 1881-1883, 1996
Transfer of gametes into the Fallopian tubes — is choice
of side important?
Geoffrey L.Driscoll1, John P.P.Tyler, Leon Clark
and Joel Bernstein
City West IVF, 12 Caroline St, Westmead, NSW 2145, Australia
'To whom correspondence should be addressed
The aim of this study was to define pregnancy rates for
gamete intra-Fallopian transfer (GIFT) with respect to the
ovary from which most oocytes were collected (i.e. a
gross index of ovarian stimulation) and the Fallopian
tube (ipsilateral or contralateral) into which gametes were
replaced. The only inclusion criterion was the ability to
transfer gametes into the Fallopian tubes at GIFT. No
other factors that could influence outcome were considered.
In this retrospective review of the clinic's database, no
relationship was found between pregnancy rate and placement of gametes into the Fallopian tube coincident with
the ovary from which most oocytes were collected. Thus
when performing unilateral tubal transfer at GIFT gametes
may be returned to the side most convenient to the operating
surgeon without fear of compromising pregnancy potential.
Key words: Fallopian tube/gamete intra-Fallopian transfer/
GIFT
Introduction
With improvement in success rates of assisted reproductive
techniques combined with the ability to cryopreserve surplus
embryos, many clinics have reduced the number of embryos
transferred in the cycle in which follicular stimulation has
been induced. Thus in Australia the number of embryos
transferred has reduced from four or more in 1986 (Fertility
Society of Australia, 1987) to three and usually less as
a recommendation of the Australian accrediting authority
(Reproductive Technology Accreditation Committee, 1992).
Similarly, in gamete intra-Fallopian transfer (GIFT), the
practice of transferring gametes to both Fallopian tubes, as
originally described (Asch et al., 1984; Leeton et al., 1987),
has generally been abandoned in Australia and oocytes and
spermatozoa are returned to the most accessible tube as long
as there is no perceived or known dysfunction on that side.
Most recent publications on GIFT make no note of unilateral
or bilateral transfer and those few studies that have specifically
examined unilateral transfer have relatively small data sets
(Haines and O'Shea, 1989, 1991).
The aim of this retrospective study was to test the hypothesis
that potential for pregnancy was maximal if gametes were
replaced into the Fallopian tube on the side coincident with
© European Society for Human Reproduction and Embryology
maximal ovarian stimulation, as defined by the collection of
most oocytes from that side.
Materials and methods
All treatment cycles on the City West IVFs database (including
patients who had multiple attempts) were reviewed for this study but
treatments in which oocytes were received from a donor, transferred
to both Fallopian tubes, or were collected from the Pouch of Douglas
were omitted. Similarly, women who did not elect to use ovulation
induction drugs were also excluded and because there were so few
cases (n = 12) were not used as a comparison group
Apart from these exclusions, no account was taken of other factors
known to contribute to the overall pregnancy rate, such as the
woman's age (range 22.1—48 7 years, median 32 3 years), body mass
index (range 9.3-84.7, median 21.7), the type of ovanan stimulation
used [clomiphene citrate (Clomid), Manon Merrell Dow, Frenchs
Forest, NSW, Australia; leuprorelin acetate (Lucrin), Abbott
Australasia, Kumell, NSW, Australia, or nafarelin acetate (Syneral),
Searle, Crows Nest, NSW, Australia, m combination with either
human menopausal gonadotrophin (HMG) or follicle stimulating
hormone (FSH)' number of 75 IU ampoules used ranged between
5-136 with a median of 20] or the subsequent response of the ovary
as determined by maximal oestrogen concentration in circulating
peripheral plasma (range 809-21 100 pmol/1, median 6120) or the
number of oocytes collected (left ovary: range 0-28, median 3; right
ovary: range 0-18, median 3). Similarly, different operative methods
for retrieving the oocytes (ultrasound guidance or laparoscopy), the
type of catheter used for gamete transfer and the potentially differing
techniques of the performing surgeon and embryologist were not
taken into consideration, given that these data were collected over a
6 year period The male partner's potential fertility was also ignored
as long as spermatozoa were deemed suitable Thus the inclusion
criterion for data in this study was simply the ability to transfer
gametes of 'visible' (but unknown) quality into the Fallopian tubes
at GIFT
The final database for review contained 1582 completed GIFT
attempts. These data were sorted according to the difference in oocyte
numbers collected from each ovary (with no account being taken of
the total number of oocytes collected) and then classified as either
left or right dominant Transfer at GIFT was considered ipsilateral if
performed on the dominant side or contralateral if not For the
majority of treatments (n = 1007; 63.7%), three oocytes were
transferred, thus minimizing any major effect of oocyte number on
pregnancy rate [29 (1.8%) had four oocytes returned, 432 (27.3%)
had two oocytes returned and 114 (7.2%) had only one oocyte
replaced]. Finally, in order not to skew results, data from women
who only had oocytes collected from one side were segregated as it
was not possible to differenuate on the database between poor
stimulation or the absence of an ovary on one side
Results
After data selection 199 women had oocytes collected from
only one ovary and 1383 had complete information that was
1881
GXJJriscoU et al
L minus R
20 0-
Lett Ovary Dominant
Right Ovary Dominant
-14 i
000%
50
362%
100
150
Frequancy
723%
1085%
% of Total
200
250
1446%
1808%
Figure 1. The distribution of the difference between the number of
oocytes collected from the right (R) and left (L) ovaries (negative
values mean more oocytes were collected from the right than from
the left).
available for review. The distribution of the difference in the
total number of oocytes collected from each ovary (number
collected from the left side minus number collected from the
right side) in the latter group is shown in Figure 1. Negative
values therefore correspond to more oocytes being collected
from the right ovary than the left. Thus the data have a slight
skew to right ovary predominance [with the interquartile range
being 3 (-2-1) oocytes difference, the 10th and 90th percentile
being - 4 and 3 respectively]. Only 20.8% of cases had a
difference greater than three oocytes. The range of oocytes
collected from both ovaries was 1-36 with a median collection
of 7.0 cells.
Table I summarizes the data and lists the pregnancy rates
according to difference in oocyte number when gametes are
replaced in either the ipsilateral or contralateral Fallopian tube.
Numbers in each group are expressed as the number of
pregnancies per total number of events exhibiting that
response. When the data were combined such that pregnancy
rate could be compared with transfer to either the ipsilateral
or contralateral tube to the ovary yielding most oocytes, then
180/633 became pregnant with ipsilateral transfer (28.4%)
and 153/547 became pregnant after contralateral transfer
(27.9%). Chi-squared analysis showed this difference to be
non-significant. Similarly, if only the extremes of oocyte
difference were considered (i.e. more than three oocytes
difference), again no significant difference was demonstrated
(contralateral 467164:28.0%; ipsilateral 37/123:30.1%). At the
other extreme, when no oocytes were collected from one ovary,
overall pregnancy rates were reduced (17.1%) but no significant
difference was found regarding the side of gamete transfer
(15/103 versus 19/96).
Discussion
With the near universal use of ultrasound-guided transvaginal
oocyte collection, the proportion of assisted reproductive
1882
technique cycles where supplementary laparoscopy for GIFT
is performed appears to be reducing but still accounts for
35.8% of all assisted reproductive technique treatments in
Australia (Fertility Society of Australia, 1995). This, coupled
with the option of transvaginal GIFT (Jansen and Anderson,
1993) and its continuing refinement (Woolcott and Stanger,
1994; Woolcott et al, 1994), continues to make relevant the
question posed in this paper.
This retrospective review of assisted reproductive technique data was not conducted in an attempt to study Fallopian
tube function but to provide reassurance of something hoped
to be true by practitioners of GIFT. However, pregnancy
outcome when gametes (or embryos) are placed in contralateral
Fallopian tubes to the maximally stimulated ovary has not to
the authors' knowledge been fully examined and, in the few
studies on unilateral tubal transfer, no account has been taken
of the side of transfer relative to ovarian stimulation (Haines
and O'Shea, 1991). While it is known that ectopic pregnancies
can occur on the contralateral tube to the ovulating ovary, and
that in women with premature ovarian failure the Fallopian
tube can transport gametes and pregnancies can occur when
donated oocytes are used (Olar et al, 1989), subtle interactions
between the ovulating ovary, its oocytes and Fallopian tube
may exist An indication for this has recently been suggested
by Kunz et al. (1996), when they demonstrated by hysterosalpingoscintigraphy preferential transport of spermatozoa into
the tube ipsilateral to the dominant follicle, confirming the
findings of Williams et al. (1993) that the number of spermatozoa associated with ovulation was also highest in the
ipsilateral tube.
This study also only examined a gross index of ovarian
stimulation (i.e. difference in oocyte numbers collected from
each ovary) and did not take into account many factors that
are known to influence pregnancy rate such as female age and
body mass index etc. Similarly, it is accepted that the ovary
from which the majority of oocytes were collected was not
necessarily the most 'stimulated' ovary from a biological
viewpoint and, unlike Kunz et al (1996) and Williams et al.
(1993) who related their data to a single dominant follicle,
these data have been collected from patients in whom both
ovaries underwent stimulation.
However, even when the difference in the number of oocytes
collected from each ovary was large (including the extreme
when no oocytes were collected from one side) the hypothesis
that transfer to the ipsilateral Fallopian tube would provide
the maximal pregnancy rate did not hold. Similarly, given that
variables known to affect pregnancy potential (i.e. female age)
were relatively evenly distributed across the difference in
oocyte groupings, then it was unlikely that any other subdivision of the data would change the finding.
A factor which could not be ignored in this data review was
the tendency of the operating surgeon to replace gametes into
the right Fallopian tube (1100 transfers to right, 283 to left
side, i.e. a ratio of 3.9:1). This was possibly because of each
surgeon's right-handedness and easier access because of the
descending sigmoid colon on the left side. Where oocytes
were transferred to the left Fallopian tube there were specific
reasons including easier access, better visual quality of the
Ovarian response and unilateral GIFT
Table I. Summary data for pregnancy rate according to the ovary (L = left, R = right) from which oocytes were collected and Fallopian tubes to which they
were replaced Thus L = 3 > R means three more oocytes collected from the left than from the right ovary
Ovanan 'response'
R
L
R
R
R
R
R
R
R
L
L
L
L
L
L
=
=
=
=
=
=
=
*
>
=
=
=
=
=
>
0
0
L
1> L
2 > L
3> L
4 > L
5 > L*
L Total
1> R
2 > R
3 > R
4 > R
5 > R*
R Total
Transfer to left side
Pregnancy
ratio R L
Transfer to right side
Pregnancies/
total number
% Pregnancy
rate
Pregnancies/
total number
%Pregnancy
rate
4/24
3/16
15/45
9/39
11/28
3/23
5/13
5/17
33/120
10/40
6723
16 7
188
33 3
23.1
39 3
13 0
38 5
29 4
27 5
250
26 1
35 0
22 2
30 8
27 9
16/80
11/79
39/158
41/172
37/130
33/84
10/55
26/74
147/515
41/161
32/112
20/61
7/31
20/62
120/427
20 0
139
24.7
23 8
28 5
39 2
18 2
35 1
28 5
255
28.6
32 8
22 6
32.3
28 1
7/20
2/9
8/26
33/118
1 030
0 725
0 742
1030
0 725
3 015
0473
1 194
1036
1020
1096
0.937
1018
1049
1007
•The small number of cases where oocyte number differences were > 5 (see Figure 1) have been combined with this grouping
fimbria, previous ectopic pregnancy or known pathology or
blockage on the other side. Similarly, each surgeon also had
a preferred side from which he started puncturing follicles,
but this does not appear to have disadvantaged a patient's
potential for pregnancy. It is interesting to speculate whether
returning oocytes collected from the 'dominant' ovary to the
ipsilateral Fallopian tube may have improved pregnancy rates
but since the source of the oocytes returned at GIFT was
random (oocytes were ranked according to the size of the
follicle from which they were aspirated and by visual features
which embryologists consider optimal before selection of the
'best' for return), this was not possible.
In conclusion, the results of this retrospective review of
assisted reproductive technique data from a non-randomized
approach to gamete replacement indicate that pregnancy rate
is not influenced by the choice of Fallopian tube in which
gametes are replaced at GIFT, and that the operating surgeon
may utilize the most convenient tube without fear of compromising a patient's pregnancy potential.
Acknowledgement
The authors are grateful to Mr N. Crockett of the Statistics Dept,
School of Economics, Macquane University, NSW, Australia, for
statistical assistance in preparing this manuscript
gamete intrafallopian transfer a case-controlled retrospective comparison
FemL Stenl, 59, 836-840
Kunz, G , Beil, D , Deuunger, H et aL (1996) The dynamics of rapid sperm
transport through the female genital tract evidence from vaginal sonography
of uterine peristalsis and hysterosalpingoscintigraphy. Hum. Reprod, 11,
627-632
Leeton, J , Rogers, P, Caro, C et aL (1987) A controlled study between the
use of gamete intrafallopian transfer (GIFT) and in vitro fertilisation and
embryo transfer in the management of ldiopathic and male infertility. Feml
Stenl, 48, 605-607
Olar, T T , Dickey, R P , Curole, D N and Taylor, S N (1989) Case report
pregnancies established by gamete intrafallopian transfer and pronuclearstage transfer in patients with premature ovanan failure using donated
oocytes and low-dose oral micronized oestradiol and progesterone /. In
Vitro Fert. Embryo Transf., 6, 160-163.
Reproductive Technology Accreditation Committee (1992) Guidelines for
Assisted Reproductive Techniques Units in Australia. Fertility Society
of Australia.
Williams, M , Hill, C J , Scudamore, I et al (1993) Sperm numbers and
distribution within the human Fallopian tube around ovuladon Hum.
Reprod, 8, 2019-2026
Wbolcott, R. and Stanger, J (1994) The fluid dynamics of injection variables
as they relate to transvaginal gamete intra-Fallopian transfer and tubal
embryo transfer Hum. Reprod., 9, 1670-1672
Woolcott, R , Stanger, J , Cohen, R and Silber, S (1994) Refinements in the
methodology of injection for transvaginal gamete mtra-Fallopian transfer
Hum. Reprod, 9, 1466-1468
Received on January 25, 1996; accepted on June 7, 1996
References
Asch, R., Ellsworth, L., Balmaceda, J and Wong, P (1984) Pregnancy after
translaparoscopic gamete intrafallopian transfer Lancet, 2, 1034
Fertility Society of Australia (1987) fVF and GIFT pregnancies in Australia
and New Zealand 1986 National Perinatal Statistics Unit, Sydney
Fertility Society of Australia (1995) Assisted Conception Australia and New
Zealand 1992 and 1993 National Perinatal Statistics Unit, Sydney
Haines, CJ and O'Shea, R.T (1989) Unilateral gamete intrafallopian transfer
the preferred method7 FertiL Stenl, 51, 518-519
Haines, CJ. and O'Shea, R.T (1991) The effect of unilateral versus bilateral
tubal cannulation and the number of oocytes transferred on the outcome of
gamete intrafallopian transfer. FertiL Stenl., 55, 423-425
Jansen, RP.S and Anderson, J C. (1993) Transvaginal versus laparoscopic
1883