Journal of Andrology, Vol. 33, No. 2, March/April 2012Copyright E American Society of Andrology
Varicocelectomy Does Not Impact Pregnancy Outcomes FollowingIntracytoplasmic Sperm Injection Procedures
FABIO F. PASQUALOTTO, DANIELA P. A. F. BRAGA, RITA C. S. FIGUEIRA, AMANDA S. SETTI,ASSUMPTO IACONELLI JR, AND EDSON BORGES JR
From the Fertility-Assisted Fertilization Center, Sao Paulo, and the Institute of Biotechnology, University of Caxias doSul, Rio Grande do Sul, Brazil.
There are many studies in the literature suggesting an
and group 2 (33.8 6 0.38 years; P 5 .1872). Semen volume was
acquired, apparently progressive infertility due to varicocele. In fact,
higher in group 1 (3.3 6 0.3 mL) than it was in group 2 (2.5 6 0.14; P 5
varicocelectomy has become the most commonly performed male
.0043). No differences were detected between groups 1 and 2 with
infertility surgery. Assisted reproductive technologies such as intra-
regard to sperm concentration (30.08 6 4.01 million/mL and 24.1 6
cytoplasmic sperm injection (ICSI) are also important for couples with
2.42 million/mL, respectively; P 5 .138), sperm motility (38.2% 6
male factor infertility associated with varicocele. Therefore, the aim of
2.69% and 38.7% 6 2.08%, respectively; P 5 .881), and morphology
this study was to evaluate the effect of varicocelectomy on sperm
according to Tygerberg’s strict criteria (2.6% 6 0.44% and 2.4% 6
quality and pregnancy rate with ICSI. Data were analyzed from 248
0.37%, respectively; P 5 .7202). Also, no differences were detected in
patients who had varicocele or underwent a previous varicocelectomy
the number of oocytes retrieved between group 1 (14.8 6 1.74) and
and were treated with ICSI between 2000 and 2008. Patients with
group 2 (14.9 6 1.04; P 5 .9515). The fertilization rate was higher in
varicocele were divided into two groups: men with clinical varicocele
group 1 (73.2%) than it was in group 2 (64.9%; P 5 .0377); however,
(group 1, n 5 79) and men who underwent varicocelectomy before
no differences were detected in the pregnancy rates (31.1% vs 30.9%;
ICSI (group 2, n 5 169). In all cases, female infertility was not
P 5 .9806), implantation rates (22.1% vs 17.3%; P 5 .5882), or
detected. We evaluated and compared the two groups’ semen
miscarriage rates (21.7% vs 23.9%; P 5 .8401) between groups 1 and
characteristics as defined by the World Health Organization and
2. Although a varicocelectomy should always be performed before
Tygerberg’s strict criteria: the female partner’s age; the number of
assisted reproduction is pursued, this surgery does not increase
oocytes retrieved; and the fertilization, implantation, pregnancy, and
pregnancy rates or decrease miscarriage rates following ICSI.
miscarriage rates. We used the Wilcoxon signed rank test or the
Varicocele, ICSI, spermatozoa, male infertility,
Mann-Whitney test for these analyses. No differences were detected
in the age of the female partners between group 1 (33.0 6 0.46 years)
Fertility problems occur in approximately 10% to The decrease in scrotal temperature following varicocele
15% of reproductive-age couples, due to female
ligation supports the causative role of increased
factor or male factor infertility (Witt and Lipshultz,
temperature in varicocele-related infertility (Sofikitis et
1993; Schlesinger et al, 1994; Kamal et al, 2001). Male
al, 1992). Additionally, it has been hypothesized that
factor infertility is at least partially responsible for 50%
varicoceles cause hypoxia, which may play a role in
of infertility cases. Varicocele is present in approximate-
impairing spermatogenesis in the patient with varico-
ly 10% to 20% of the healthy male population, but in
cele. Furthermore, infertile men with varicocele have
30% to 40% of the infertile male population (Witt and
elevated levels of spermatozoal reactive oxygen species
Lipshultz, 1993; Schlesinger et al, 1994; Kamal et al,
(Hendin et al, 1999; Pasqualotto et al, 2000, 2008;
2001). A number of theories have been proposed to
explain the decreased semen quality detected in infertile
The effects of varicocele vary, but they often result in
men with varicocele. Semen quality uniformly declines
an overall impairment of sperm production character-
in animals with induced varicoceles, even when only a
ized by abnormal semen quality and a reduced fertilizing
left varicocele is present (Sofikitis and Miyagawa, 1994).
capacity of the male gamete (Sofikitis et al, 1996; Gat etal, 2003, 2004). Currently, surgical repair of varicocele isreported to be more cost-effective than intracytoplasmic
Correspondence to: Edson Borges Jr, Fertility-Assisted Fertilization
sperm injection (ICSI; Garceau et al, 2002; Meng et al,
Center, Av Brigadeiro Luı´s Antoˆnio, Sa˜o Paulo, SP 01401-002, Brazil
2005). However, even when varicocele is repaired and all
(e-mail: [email protected]
beneficial improvements in semen quality appear to
Received for publication September 15, 2010; accepted for
have been obtained, unassisted pregnancy rates vary
from 19% to 35% (Marmar and Kim, 1994; Gat et al,
2003, 2004; Pasqualotto et al, 2006), and these rates are
according to WHO criteria (Cooper et al, 2010), and
significantly lower in patients who had germ cell aplasia
morphology according to Tygerberg’s strict criteria (Kruger
before surgery (Pasqualotto et al, 2005).
et al, 1986). At least two centrifuged semen samples were
Varicocele repair has induced or enhanced spermato-
genesis in 40% to 60% of men with azoospermia and
severe oligoasthenozoospermia (Matthews et al, 1998;Kim et al, 1999; Kadioglu et al, 2001; Pasqualotto et al,
The presence of granulocytes in the semen specimens was
2006; Inci et al, 2009). However, few pregnancies have
assessed by a myeloperoxidase test (Kruger et al, 1986). A 20-mL
been reported to result after varicocelectomy in men
volume of liquefied specimen was placed in a Corning 2.0-mLcryogenic vial (Corning Costar Corp, Cambridge, Massachu-
with varicocele and azoospermia. Therefore, ICSI is a
setts), to which 20 mL of phosphate-buffered saline (pH 7.0) and
valuable tool for couples in which the male partner has
40 mL of benzidine solution were added. The mixture was
varicocele and azoospermia, because not all couples
vortexed and allowed to sit for 5 minutes. A 5-mL portion of the
achieve pregnancy after varicocele repair alone.
specimen was placed on a Makler chamber (Sefi Medical, Haifa,
The aim of this study was to evaluate the effect of
Israel) and examined for cells that stained dark brown, which
varicocelectomy on sperm quality and pregnancy rate
indicated that they were positive for neutrophils. Leukocyto-
spermia was defined as the presence of at least 1 6 106 whiteblood cells per milliliter. In our study, we excluded patients whohad more than 1.0 6 106 white blood cells per milliliter.
The patients’ female partners completed an infertility workup
This study was approved by the University of Caxias do Sul
consisting of at least one evaluation assessment and hystero-
review board, and the patients involved gave informed
salpingography. Female partners’ mean age 6 SD was 30.6 6
consent. A retrospective study of 248 patients who had
varicocele or had previously undergone a varicocelectomy,
The patients’ female partners underwent a unique controlled
and who were treated with ICSI between 2000 and 2008, was
ovarian stimulation protocol. Gonadotropin-releasing hor-
performed. Only patients presenting varicocele grade III were
mone agonist (leuprolide acetate) was administered for at least
14 days. Ovarian stimulation with recombinant follicle-
For the purposes of this study, infertility was defined as the
stimulating hormone (Gonal F; Merck Serono Laboratories,
failure to establish a pregnancy after at least 1 year of
Geneva, Switzerland) was initiated in a step-down protocol
unprotected intercourse. A basic infertility evaluation, including
until a minimum of 2 follicles reached an average diameter of
a detailed history and a complete physical examination, was
18 mm. Oocyte retrieval was performed using transvaginal
conducted. Only patients who were diagnosed with varicocele
ultrasonography 36 hours after the administration of 10 000 IU
grade III were submitted to receive assisted reproduction
of human chorionic gonadotropin. The oocytes were kept in
treatment or varicocelectomy. In addition, semen analyses were
human tube fluid medium supplemented with 7.5% synthetic
performed in all patients, and only those presenting poor semen
serum substitute for approximately 3 to 5 hours before
parameters according to World Health Organization (WHO)
cumulus cell removal. The cumulus cells were removed from
2010 guidelines were included. All couples participating in the
the oocytes by placing them in hyaluronidase at a concentra-
study underwent ICSI cycles for the first time.
tion of 80 IU/mL for approximately 30 to 60 seconds. The
Patients who had varicocelectomy and recurrence of the
oocytes were transferred to fresh medium, and the corona cells
varicocele were excluded from the study. Patients who were
taking antioxidants, such as vitamins C and E, and those whohad azoospermia or leukocytospermia were also excluded from
ICSI was performed with oocytes in metaphase II (MII),
Patients with varicocele were divided into two groups: men
following the technique described by Palermo et al (1992). The
with clinical varicocele (group 1, n 5 79) and men who had
microinjection dish was prepared by pipetting seven 4-mL
undergone varicocelectomy before ICSI (group 2, n 5 169).
droplets of human tubular fluid medium and adding HEPES
All 248 patients underwent a subinguinal varicocelectomy with
and 15% synthetic serum substitute to the center of the dish.
magnification, as previously described (Marmar and Kim,
The center droplet was removed and replaced with 4 mL of
polyvinylpyrrolidone solution. Prepared sperm was added tothe drop of polyvinylpyrrolidone solution and the oocytes, and
then placed into the drops of medium. The droplets were
Semen samples were obtained by masturbation after at least
overlaid with mineral oil. ICSI was performed microscopically
48 hours of abstinence. Samples were collected into sterile
on a heated stage. After the oocytes were injected, they were
containers and allowed to liquefy at 37uC for 30 minutes and
rinsed and placed in equilibrated growing medium (15%
then analyzed for sperm concentration, percent motility
human tubular fluid–synthetic serum substitute). The micro-
Varicocelectomy and Assisted Reproductive Outcomes
Table. Parameters evaluated in patients with varicocele (n 5 79) and patients who underwent a previous varicocelectomy (n5 169)
Abbreviations: FSH, follicle-stimulating hormone; MII, metaphase II.
manipulated oocytes were incubated in 50- to 100-mL drops in
The couples in group 2 had had tried longer to achieve a
a culture dish overlaid with mineral oil. The oocytes were
pregnancy (6.0 6 0.49 years) than those in group 1 (2.7
observed for fertilization for 16 to 18 hours after the injection
procedure. Abnormally fertilized or unfertilized oocytes were
Semen volume was higher in group 1 (3.3 6 0.3 mL)
removed from the dish. Twenty-four hours later, cleavage of
than in group 2 (2.5 6 0.14 mL; P 5 .0043). However,
fertilized oocytes was assessed. Embryo transfer was per-formed approximately 72 hours after injection.
no differences were detected between groups 1 and 2 insperm concentration (30.08 6 4.01 million/mL and 24.1
6 2.42 million/mL; P 5 .138), sperm motility (38.2% 62.69% and 38.7% 6 2.08%; P 5 .881), and morphology
Clinical pregnancy was determined by the visualization of atleast one gestational sac by ultrasound at 7 weeks.
according to Tygerberg’s strict criteria (2.6% 6 0.44%and 2.4% 6 0.37%; P 5 .7202). The mean volume (in
milliliters) of the left testicle did not differ betweengroup 1 (19.4 6 0.71) and group 2 (18.8 6 0.66; P 5
We compared the age of the male patients with their femalepartners, the men’s semen characteristics (according to WHO
.4929), nor did the mean volume of the right testicle
and Tygerberg’s strict criteria), right and left testicular volume,
(20.3 6 0.60 and 19.9 6 0.67 for groups 1 and 2,
the women’s follicle-stimulating hormone and estradiol levels,
the number of oocytes retrieved, the number of MII oocytes
The mean follicle-stimulating hormone levels of the
retrieved, the number of cycles canceled, fertilization, implan-
female partners in both groups were not different (2371.7
tation, the number of embryos transferred, and pregnancy and
6 100.69 IU and 2249.4 6 63.68 IU for groups 1 and 2,
miscarriage rates between the two groups. For this analysis, we
respectively; P 5 .2943), nor were mean estradiol levels
used the Wilcoxon signed rank test or the Mann-Whitney test.
(1926.8 6 364.73 pg/mL and 2002.4 6 201.33 pg/mL forgroups 1 and 2, respectively; P 5 .8490).
There were no differences in the number of oocytes
retrieved in group 1 (14.8 6 1.74) and group 2 (14.9 61.04; P 5 .9515), and there was no difference in the
No differences were detected in the age of the men in
percentage of oocytes retrieved (65.4% for group 1 and
group 1 (36.1 6 0.55 years) and group 2 (37.8 6
67.6% for group 2; P 5 .6401). No differences were
0.47 years; P 5 .1872; Table), and no differences were
detected in the mean of number of MII oocytes retrieved
detected in the age of the women in group 1 (33.0 6
in group 1 (8.8 6 1.34) and in group 2 (10.4 6 0.79; P 5
0.46 years) and group 2 (33.8 6 0.38 years; P 5 .1872).
.2805). There was no difference in the mean number of
oocytes injected per couple in group 1 (10.3 6 1.16) and
peutic approach for infertile couples with varicocele is
whether to treat this condition before using TESE/ICSI.
The fertilization rate was higher in group 1 (73.2%)
Treating varicocele is more cost-effective than other
than in group 2 (64.9%; P 5 .0377). However, no
infertility treatments, especially if future pregnancies are
differences were detected in the pregnancy rates (31.1%
planned (Van Steirteghem et al, 1993; Penson et al,
vs 30.9%; P 5 .9806), implantation rates (22.1% vs
2002). On the other hand, Lee et al (2009) demonstrated
17.3%; P 5 .5882), and miscarriage rates (21.7% vs
that when direct costs are considered, microsurgical
23.9%; P 5 .8401) in groups 1 and 2. No differences
TESE appears to be more cost-effective than varicocel-
were detected in the mean number of embryos trans-
ectomy for treating varicocele-associated nonobstructive
ferred in group 1 (2.8 6 0.18) and group 2 (2.4 6 0.13; P
azoospermia (Haydardedeoglu et al, 2010). The cost-
5 .0785), or in the number of cycles cancelled per cycle
effectiveness of both treatments has improved with time.
(5.1% and 10.1%, respectively; P 5 .2277).
Most importantly, the authors concluded that cost-effectiveness analyses may be tailored to institution-specific data to allow more individualized results.
A retrospective study by Haydardedeoglu et al (2010)
evaluated whether a history of varicocele repair in men
Varicocelectomy has become the most commonly
with nonobstructive azoospermia improved ICSI out-
performed male infertility surgery. In fact, according
comes. The authors found a significantly higher sperm
to Zini et al (2008), men with poorer baseline
retrieval rate in the varicocele repair group compared
characteristics are more likely to opt for varicocele
with the group of men who had not undergone
repair. Furthermore, couples electing not to repair the
varicocele repair. The clinical pregnancy and live birth
varicocele are more likely to undergo ICSI procedures to
rates were significantly higher in the varicocelectomy
improve their chances of conception.
group. Therefore, varicocele repair might be considered
The relationship between varicocele and azoospermia
in patients with nonobstructive azoospermia who are
is controversial. The main question is whether azoo-
spermia is caused by varicocele or by coexisting primary
Smit et al (2010) in a very elegant study prospectively
testicular failure. Therefore, there is no consensus on
evaluated sperm chromatin structure changes in infertile
whether to correct varicocele in azoospermic men with
patients before and after surgical varicocele repair and
maturation arrest or Sertoli cell–only syndrome. Fur-
their impact on pregnancy rates. After varicocelectomy,
thermore, the question remains whether surgical repair
sperm parameters significantly improved and sperm DNA
of varicocele should be recommended for the above
fragmentation was significantly decreased. Low DNA
conditions to avoid the need to obtain sperm by
fragmentation index values were associated with a higher
testicular sperm extraction (TESE) for ICSI. Motile
pregnancy rate (both spontaneous and with assisted
sperm from the ejaculate provide better fertilization,
reproductive techniques). Therefore, the authors argued
cleavage, and pregnancy rates with ICSI than those
that varicocelectomy should be considered in infertile men
from TESE (Loutradi et al, 2006). It is generally
with palpable varicocele, men with abnormal semen
accepted that sperm production recovers somewhat
analysis, and female partners with no major infertility
after varicocele repair, but remains suboptimal. In a
factors. However, the study included only 49 men, which
recent study, Inci et al (2009) evaluated 96 men with
was too small a sample to allow for conclusive results.
complete nonobstructive azoospermia and a history ofclinical unilateral or bilateral varicocele. Their results
In our study of 248 patients who had varicocele or
suggest that varicocele repair significantly increases the
who had previously undergone a varicocelectomy, we
sperm retrieval rate in patients with clinical varicocele
found that although the fertilization rate was higher in
and nonobstructive azoospermia. The authors conclud-
the varicocele group, the pregnancy, implantation, and
ed that men with clinical varicocele and nonobstructive
miscarriage rates did not differ between the groups.
azoospermia who undergo varicocelectomy are not
Further studies are needed to confirm our results.
likely to retain sufficient sperm quality to achieve
pregnancy either with intercourse, intrauterine insemi-nation, or ICSI without the use of TESE.
We conclude that although past evidence indicates that
The development of ICSI as the method of choice in
varicocelectomy may be of benefit in some circumstanc-
cases of severe male factor infertility has encouraged
es and should always be considered before recommend-
scientists to modify their diagnostic and therapeutic
ing assisted reproductive techniques when appropriate,
approaches (Van Steirteghem et al, 1993; Penson et al,
this surgical procedure was not shown to have an impact
2002). Thus, a frequent question regarding the thera-
on pregnancy or miscarriage rates following ICSI in
Varicocelectomy and Assisted Reproductive Outcomes
couples where the male partner had an existing or
Matthews GJ, Matthews ED, Goldstein M. Induction of spermato-
genesis and achievement of pregnancy after microsurgical varico-celectomy in men with azoospermia and severe oligoasthenosper-mia. Fertil Steril. 1998;70:71–75.
Meng MV, Greene KL, Turek PJ. Surgery or assisted reproduction?: a
decision analysis of treatment costs in male infertility. J Urol.
2005;174:1926–1931; discussion 1931.
Cooper TG, Noonan E, von Eckardstein S, Auger J, Baker HW, Behre
Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies
HM, Haugen TB, Kruger T, Wang C, Mbizvo MT, Vogelsong
after intracytoplasmic injection of single spermatozoon into an
KM. World Health Organization reference values for human
semen characteristics. Hum Reprod Update. 2010;16:231–245.
Pasqualotto FF, Lucon AM, de Goes PM, Hallak J, Sobreiro B,
Garceau L, Henderson J, Davis LJ, Petrou S, Henderson LR,
Pasqualotto EB, Arap S. Testicular growth, sperm concentration,
McVeigh E, Barlow DH, Davidson LL. Economic implications
percent motility, and pregnancy outcome after varicocelectomy
of assisted reproductive techniques: a systematic review. Hum
based on testicular histology. Fertil Steril. 2005;83:362–366.
Gat Y, Bachar GN, Zukerman Z, Belenky A, Gornish M. Varicocele:
Pasqualotto FF, Sharma RK, Nelson DR, Thomas AJ, Agarwal A.
a bilateral disease. Fertil Steril. 2004;81:424–429.
Relationship between oxidative stress, semen characteristics, and
Gat Y, Zukerman ZV, Bachar GN, Feldberg DO, Gornish M.
clinical diagnosis in men undergoing infertility investigation. Fertil
Adolescent varicocele: is it a unilateral disease? Urology. 2003;62:
Pasqualotto FF, Sobreiro BP, Hallak J, Pasqualotto EB, Lucon AM.
Haydardedeoglu B, Turunc T, Kilicdag EB, Gul U, Bagis T. The effect
Induction of spermatogenesis in azoospermic men after varicocel-
of prior varicocelectomy in patients with nonobstructive azoosper-
ectomy repair: an update. Fertil Steril. 2006;85:635–639.
mia on intracytoplasmic sperm injection outcomes: a retrospective
Pasqualotto FF, Sundaram A, Sharma RK, Borges E Jr, Pasqualotto
pilot study. Urology. 2010;75:83–86.
EB, Agarwal A. Semen quality and oxidative stress scores in
Hendin BN, Kolettis PN, Sharma RK, Thomas AJ Jr, Agarwal A.
fertile and infertile patients with varicocele. Fertil Steril. 2008;
Varicocele is associated with elevated spermatozoal reactive oxygen
species production and diminished seminal plasma antioxidant
Penson DF, Paltiel AD, Krumholz HM, Palter S. The cost-
capacity. J Urol. 1999;161:1831–1834.
effectiveness of treatment for varicocele related infertility. J Urol.
Inci K, Hascicek M, Kara O, Dikmen AV, Gurgan T, Ergen A. Sperm
retrieval and intracytoplasmic sperm injection in men with
Schlesinger MH, Wilets IF, Nagler HM. Treatment outcome after
nonobstructive azoospermia, and treated and untreated varicocele.
varicocelectomy: a critical analysis. Urol Clin North Am. 1994;21:
Kadioglu A, Tefekli A, Cayan S, Kandirali E, Erdemir F, Tellaloglu S.
Smit M, Romijn JC, Wildhagen MF, Veldhoven JL, Weber RF, Dohle
Microsurgical inguinal varicocele repair in azoospermic men.
GR. Decreased sperm DNA fragmentation after surgical varico-
celectomy is associated with increased pregnancy rate. J Urol.
Kamal KM, Jarvi K, Zini A. Microsurgical varicocelectomy in the era
of assisted reproductive technology: influence of initial semen
Sofikitis N, Miyagawa I. Bilateral effect of unilateral varicocele on
quality on pregnancy rates. Fertil Steril. 2001;75:1013–1016.
testicular metabolism in the rabbit. Int J Fertil Menopausal Stud.
Kim ED, Leibman BB, Grinblat DM, Lipshultz LI. Varicocele repair
improves semen parameters in azoospermic men with spermato-
Sofikitis N, Takahashi C, Nakamura I, Hirakawa S, Miyagawa I.
genic failure. J Urol. 1999;162:737–740.
Surgical repair of secondary right varicocele in rats with primary
Kruger TF, Menkveld R, Stander FS, Lombard CJ, Van der Merwe
left varicocele: effects on fertility, testicular temperature, sper-
JP, van Zyl JA, Smith K. Sperm morphologic features as a
matogenesis, and sperm maturation. Arch Androl. 1992;28:43–52.
prognostic factor in in vitro fertilization. Fertil Steril. 1986;46:
Sofikitis NV, Miyagawa I, Incze P, Andrighetti S. Detrimental effect of
left varicocele on the reproductive capacity of the early haploid
Lee R, Li PS, Goldstein M, Schattman G, Schlegel PN. A decision
male gamete. J Urol. 1996;156:267–270.
analysis of treatments for nonobstructive azoospermia associated
Van Steirteghem AC, Nagy Z, Joris H, Liu J, Staessen C, Smitz J,
with varicocele. Fertil Steril. 2009;92:188–196.
Wisanto A, Devroey P. High fertilization and implantation rates
Loutradi KE, Tarlatzis BC, Goulis DG, Zepiridis L, Pagou T,
after intracytoplasmic sperm injection. Hum Reprod. 1993;8:
Chatziioannou E, Grimbizis GF, Papadimas I, Bontis I. The effects
of sperm quality on embryo development after intracytoplasmic
Witt MA, Lipshultz LI. Varicocele: a progressive or static lesion?
sperm injection. J Assist Reprod Genet. 2006;23:69–74.
Marmar JL, Kim Y. Subinguinal microsurgical varicocelectomy: a
Zini A, Boman J, Baazeem A, Jarvi K, Libman J. Natural history of
technical critique and statistical analysis of semen and pregnancy
varicocele management in the era of intracytoplasmic sperm
injection. Fertil Steril. 2008;90:2251–2256.
Is Transcending Natural Boundaries Ethical? Reservation: Free admission , but reservation required. Please reserve your seats with Mr Wilson Wu at 6773 6475 or [email protected] with your name, institution / organisation and contact details by 26 July 2010 (Monday) Maxwell Auditorium, Science Centre Singapore Organisers: Centre for Biomedical Ethics, National Un
BRAVO Capsule for Esophageal pH Monitoring About the Test : BRAVO pH Test is performed to test for gastro-esophageal reflux disease (GERD). BRAVO is a catheter free pH monitoring device that assists in evaluating the frequency and severity of your acid reflux disease. The BRAVO capsule is placed in the lower or upper esophagus; this may be done at the time of an upper endoscopy (EGD