National study of factors influencing assisted reproductive ...

National study of factors influencing assisted
reproductive technology outcomes with male factor
infertility
Ajay K. Nangia, M.B.B.S., a Barbara Luke, Sc.D., M.P.H., b James F. Smith, M.D., M.S., c
Winifred Mak, M.D., Ph.D., d Judy E. Stern, Ph.D., e and the SART Writing Group
a Department of Urology, University of Kansas Medical Center, Kansas City, Kansas; b Department of Urology, Michigan State
University, East Lansing, Michigan; c Department of Urology, University of California, San Francisco, San Francisco, California;
d Department of Obstetrics and Gynecology, University of Pennsylvania, Philadelphia, Pennsylvania; and e Department of
Obstetrics and Gynecology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
Objective: To evaluate the outcomes of assisted reproductive technology (ART) cycles for male factor infertility,
and method of sperm collection.
Design: Historic cohort study.
Setting: Clinic-based data.
Patients: Cycles from the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System database
for 2004 to 2008 were limited to three groups: non-intracytoplasmic sperm injection (ICSI) and ICSI cycles
for tubal ligation only; non-ICSI and ICSI cycles for male factor infertility only; and all cycles (regardless of
infertility diagnosis) using ICSI only.
Intervention(s) and Main Outcome Measure(s): Multivariate logistic regression was used to model the adjusted odds
ratio (AOR) of clinical intrauterine gestation (CIG) and live birth (LB) rates for tubal ligation versus male factor infertility
only; ICSI versus non-ICSI for male factor infertility only; and ICSI outcomes based on method of sperm collection.
Result(s): Models for male factor infertility only versus tubal ligation only ICSI cycles had lower CIG (AOR 0.92)
but not LB (AOR 0.87). No difference was seen for non-ICSI cycles. Within male factor infertility only cycles, ICSI
had a worse outcome than non-ICSI for CIG (AOR 0.93) but not for LB (AOR 0.94). For all ICSI cycles with no
male factor infertility and ejaculated sperm as the reference group, models showed better rates of CIG with male
factor infertility ejaculated sperm (AOR 1.07) and with male factor infertility aspirated sperm (AOR 1.09). The LB
rate was higher with male factor infertility ejaculated sperm only (AOR 1.04).
Conclusion(s): The ICSI and sperm source influence CIG and LB rates in male factor infertility cases. (Fertil
Steril Ò 2011;96:609–14. Ó2011 by American Society for Reproductive Medicine.)
Key Words: SART, male factor infertility, ICSI, sperm collection methods
At least 50% of infertility is partly or completely attributable to
a male factor (1). A Centers for Disease Control and Prevention
(CDC) study of men in the 2002 National Survey for Family Growth
found that 7.5% of all sexually experienced men reported a visit for
help with having a child at some time during their lifetime (equating
to 3.3–4.7 million men) (2). Of men who sought help, 13.7% were
diagnosed with a sperm or semen problem (2). Reduced fertilization
rates, progression to blastocyst, pregnancy, and live birth outcomes
from assisted reproductive technology (ART), especially intracytoplasmic
sperm injection (ICSI), have been shown to be related to the
severity of male factor infertility by individual groups but has not
been studied at a national level (3–7). Similarly, the outcomes
based on method of sperm acquisition have been limited and
variable (8, 9). This is largely due to there being no formal
standardized national database regarding the definitions and
prevalence of male factor infertility or outcomes.
Received April 17, 2011; revised June 8, 2011; accepted June 9, 2011;
published online July 5, 2011.
A.K.N. has nothing to disclose. B.L. has nothing to disclose. J.F.S. has
nothing to disclose. W.M. has nothing to disclose. J.E.S. has nothing
to disclose.
Reprint requests: Ajay K. Nangia, M.B.B.S., Department of Urology, University
of Kansas Medical Center, MS3016, 3901 Rainbow Boulevard,
Kansas City, Kansas 66160 (E-mail: anangia@kumc.edu).
At present databases based on insurance (e.g., Kaiser or Prudential),
ambulatory medicine and surgery usage, Veterans Administration
coverage, and inpatient care by the Health Care and Utilization
Project provide limited information (10). The National Survey of
Family Growth and ART outcomes collected by Society for Assisted
Reproductive Technology (SART) and the CDC currently serve as
the only method to track infertility demographics and treatments
on a national level (11, 12). More than 140,000 ART cycles are
reported from more than 350 clinics. Information on these cycles
is entered into the SART and CDC databases under the
requirements of the Fertility Clinic Success Rate and Certification
Act of 1992 (13). Many parameters associated with each cycle are
reported by each individual center and referred to as the SART
Clinic Outcomes Reporting System (CORS).
Male factor infertility is reported by SART CORS but only as
a single diagnosis without subdividing into etiologies. The method
of sperm acquisition used for the cycle is also recorded. On a national
level, no previous report of the outcomes of ART cycles due to male
factor infertility or based on the method of sperm acquisition has
been performed. The aims of our study were to determine the effect
of male factor infertility on conventional IVF (non-ICSI) and ICSI
clinical outcomes compared with a control group (tubal ligation
only cycles); to determine the difference between outcomes from
ICSI and non-ICSI for male factor infertility only cycles; and to
0015-0282/$36.00 Fertility and Sterility â Vol. 96, No. 3, September 2011 609
doi:10.1016/j.fertnstert.2011.06.026 Copyright ª2011 American Society for Reproductive Medicine, Published by Elsevier Inc.

TABLE 1
Distribution of parameters for conventional IVF (non-ICSI) versus ICSI in male factor infertility cycles versus tubal ligation only.
ICSI—none
ICSI
Male factor only No male factor Male factor only No male factor
No female
factor
Tubal
ligation only
P value
No female
factor
Tubal
ligation only
P value
No. of cycles 4,973 3,649 72,459 3,349
Female age (%)

TABLE 2
Models for male factor only versus tubal ligation only using conventional IVF (non-ICSI) or ICSI.
Outcome
Treatment outcome
Clinical intrauterine
pregnancy vs. not
pregnant or other a
Pregnancy outcome
Live birth vs. fetal death
or stillbirth
Male factor infertility
groups
No male factor, tubal
ligation only
Male factor only, no
female factor
No male factor, tubal
ligation only
Male factor only, no
female factor
Conventional IVF (non-ICSI)
ICSI
AOR 95% CI P value AOR 95% CI P value
1.00 Reference 1.00 Reference
1.05 0.94–1.16 .40 0.92 0.85–0.99 .03
1.00 Reference 1.00 Reference
1.02 0.84–1.25 .83 0.87 0.75–1.02 .08
Note: Models adjusted for woman’s age, male and female race/ethnicity, day of ET, and number of embryos transferred. In all groups, only ejaculated sperm
was used. ICSI ¼ intracytoplasmic sperm injection; AOR ¼ adjusted odds ratio; CI ¼ confidence interval.
a Other includes biochemical, ectopic, and heterotopic.
Nangia. Male factor infertility and ART outcomes in the USA. Fertil Steril 2011.
performed. The second comparison was to model and compare the outcomes
from non-ICSI cycles versus ICSI within the male factor infertility only
group. The final analysis was to compare the outcomes from ICSI male factor
infertility cycles by the different methods of sperm acquisition versus
non-male factor infertility cycles using ejaculated sperm. All models were
adjusted for female age; male and female race/ethnicity; day of ET; and number
of embryos transferred. Models for the sperm acquisition comparison
were additionally adjusted for female factor infertility diagnoses. Results
are presented as adjusted odds ratios (AOR) and 95% confidence intervals.
RESULTS
The total number of fresh autologous cycles of IVF performed between
2004 and 2008 was 465,046. Of these, 272,526 cycles
(58.6%) were ICSI. Tubal ligation without any other factor accounted
for 6,998 IVF cycles. This represents 1.5% of all IVF cycles
performed during the 5-year period. Of these 3,649 were non-ICSI
cycles (52%) and 3,349 involved ICSI (48%). A total of 77,432
IVF cycles were attributed to male factor infertility only. These accounted
for 16.7% of all IVF cycles performed during the 5-year period
studied. Of these, 4,973 (6.4%) cycles were non-ICSI and
72,459 (93.6%) used ICSI.
Of the 272,526 ICSI cycles, 136,547 (50.1%) were performed for
female factor only using ejaculated sperm; 63,520 cycles (23.3%)
for combined male and female factor infertility; and 72,459
(26.6%) for male factor infertility only. In all, 135,979 cycles
(49.9%) had a male factor infertility component. Stratification of
these ICSI cycles with a male factor infertility component by
method of sperm acquisition showed use of ejaculated sperm in
121,626 cycles (89.4%); biopsy in 6,986 cycles (5.1%); sperm aspiration
in 6,934 cycles (5%); retrograde ejaculation in 187 cycles
(0.13%); and electroejaculation in 246 cycles (0.18%).
The first comparison was to determine the difference in outcomes
between tubal ligation only and male factor infertility only cycles.
Non-ICSI and ICSI cycles were compared separately. Groups differed
significantly in maternal age and day of ET (Table 1). For
non-ICSI, male factor infertility only cycles compared with tubal ligation
only cycles had a significantly lower CIG rate; longer length
of gestation and higher birthweights for singleton and twin pregnancies.
The LB rate did not differ. However, after adjustment for female
age, male and female race/ethnicity, day of ET, and number
of embryos transferred, models showed no significant difference
in the AOR for CIG or LB for the male factor infertility only cycles
TABLE 3
Model for treatment outcomes from conventional IVF (non-ICSI) versus ICSI cycles due to male factor infertility only.
Male factor only—no female factor
Outcome
Use of ICSI
AOR 95% CI P value
Treatment outcome
Clinical intrauterine pregnancy
ICSI—none 1.00 Reference
vs. not pregnant or other a ICSI—some or all 0.93 0.87–0.99 .03
Pregnancy outcome
Live birth vs. fetal death or
stillbirth
ICSI—none 1.00 Reference
ICSI—some or all 0.91 0.79–1.04 .16
Note: Model adjusted for woman’s age, male and female race/ethnicity, day of ET, and number of embryos transferred. In all groups, only ejaculated sperm
was used. ICSI ¼ intracytoplasmic sperm injection; AOR ¼ adjusted odds ratio; CI ¼ confidence interval.
a Other includes biochemical, ectopic, and heterotopic.
Nangia. Male factor infertility and ART outcomes in the USA. Fertil Steril 2011.
Fertility and Sterility â 611

TABLE 4
Models of treatment outcomes and pregnancy outcomes by male factor infertility and semen method of collection (all ICSI).
Dependent Independent AOR 95% CI P value
Treatment outcome
Clinical intrauterine pregnancy No male factor, ejaculated 1.00 Reference
Male factor, ejaculated 1.07 1.05–1.09 < .0001
Male factor, biopsy 1.03 0.97–1.08 .38
Male factor, aspiration 1.09 1.03–1.15 .002
Male factor, retrograde 0.85 0.62–1.16 .30
Male factor, electroejaculation 0.95 0.73–1.25 .73
Pregnancy outcome
Live birth No male factor, ejaculated 1.00 Reference
Male factor, ejaculated 1.04 1.00–1.09 .05
Male factor, biopsy 1.07 0.95–1.19 .27
Male factor, aspiration 1.07 0.96–1.20 .22
Male factor, retrograde 0.63 0.36–1.12 .12
Male factor, electroejaculation 1.02 0.59–1.75 .96
Note: Models adjusted for woman’s age, male and female race/ethnicity, day of ET, number of embryos transferred, female diagnoses of endometriosis,
ovulation disorders, diminished ovarian reserve, tubal ligation, tubal hydrosalphinx, other tubal factors, uterine factors, other factors, and unexplained
factors.
Nangia. Male factor infertility and ART outcomes in the USA. Fertil Steril 2011.
compared with the tubal ligation only non-ICSI reference group
(Table 2).
In the ICSI cohorts from the tubal ligation and male factor infertility
only groups, CIG, LB, length of gestation, and birthweights
were significantly better with male factor infertility cycles versus
tubal ligation. However, correcting for variables, the model showed
that CIG rate was significantly lower in the male factor infertility
group compared with the tubal ligation reference group. The LB
rates did not differ significantly, although they did show a trend
toward being lower (P¼.08) (Table 2).
The next comparison performed was to determine the outcomes
from non-ICSI versus ICSI within the male factor infertility only
group with ejaculated sperm. Non-ICSI and ICSI cycles differed significantly
by maternal age, male and female race/ethnicity. After adjustments
for female age, male and female race/ethnicity, day of ET,
number of embryos transferred, and with non-ICSI cycles as the reference
group, models showed ICSI to have a lower CIG rate, but not
a lower LB rate (Table 3).
Outcomes by sperm acquisition method for ICSI were divided
into cycles with no male factor infertility that used ejaculated sperm
and cycles with male factor, either alone or in combination with female
factor infertility, using sperm collected by the different
methods. Models were adjusted for female age, male and female
race/ethnicity, day of ET, number of embryos transferred, and female
diagnoses. Using non-male factor infertility ejaculated sperm
as the reference group, models for ICSI had significantly better rates
of CIG and LB with male factor infertility ejaculated sperm. Male
factor infertility aspirated sperm also had a CIG rate that was significantly
better than the reference group, but showed no difference in
LB rate (Table 4). Sperm from biopsy, retrograde ejaculation, and
electroejaculation were not statistically different from the nonmale
factor infertility cycles (Table 4).
DISCUSSION
Before 1992, conventional IVF could not address many issues related
to male factor infertility and relied on normal or near-normal
sperm counts. The severity of male factor infertility based on semen
analysis data has been partly managed with the introduction of ICSI.
The use of ICSI has revolutionized the field with regard to severe
male infertility but outcomes from such male factor infertility cases
have not been studied using a national database.
Our study showed that male factor infertility alone accounted for
16.7% of all fresh autologous ART cycles in the United States during
a 5-year period. The majority of male factor infertility only cycles
used ICSI (93.6%). Of all ICSI cycles performed during the 5-
year period, more than half had a male factor component, either
mixed male-female factor infertility or pure male factor infertility,
which continues to support previous literature (1). Ejaculated sperm
was the most often used sperm source for male factor infertility ICSI
cases (89.4%). Because the majority of male factor infertility cycles
involve ICSI would imply a need for ICSI based on significantly abnormal
sperm parameters. However, the ICSI cycles performed do
not necessarily indicate severity of the male factor infertility, as
this could not be determined from the limited definition of the
data point ‘‘male factor’’ as found in SART CORS. No indication
of average sperm count or stratification by semen analysis is given.
The reason for the use of ICSI cannot be determined specifically
from SART CORS.
Our study demonstrated that ICSI cycles for male factor infertility
compared with tubal ligation had a lower outcome for CIG and
a trend toward lower LB rates. This was not seen with non-ICSI.
Length of gestation and birthweights for the LBs were not affected
by male factor infertility. A possible relationship between male factor
infertility and worse CIG outcomes with ICSI may again be related
to sperm dysfunction and not the in vitro technique itself, with
tubal ligation only cycles chosen as a reference group to act specifically
as the control for the iatrogenesis of the technique. The fact
that non-ICSI cycles for male factor infertility showed no difference
in outcomes from the tubal ligation only group may suggest that
milder male factor infertility, which did not warrant ICSI, has little
to no effect on ART outcomes. This is an inherent bias in the data to
study the effect of male factor infertility. Live birth rate, however,
does not exclude birth defects or subsequent risks that cannot be
studied from SART CORS. Other studies have indicated that severity
of male factor infertility may be associated with worsened
clinical outcomes including failed fertilization, poor embryo
612 Nangia et al. Male factor infertility and ART outcomes in the USA Vol. 96, No. 3, September 2011

progression, miscarriages, and genetic/epigenetic inheritance abnormalities
(3–7, 14–19). Our study demonstrated that ICSI for male
factor infertility had lower CIG rates than non-ICSI IVF for male
factor infertility. This finding does not categorically indicate that
outcomes are worse due to technique (i.e., ICSI versus non-ICSI)
or severity of the male factor infertility requiring ICSI, as the data
cannot be stratified by severity. Also, only ejaculated sperm used
in male factor infertility only ICSI and non-ICSI cycles was considered
to allow for a matched comparison. There were insufficient
numbers of male factor infertility only cycles for ICSI due to other
forms of sperm acquisition, especially biopsy and aspiration to substratify
the comparison. Use of ICSI for male factor infertility can
only imply severity of the male infertility.
Many investigators have explored the relationship between sperm
acquisition source and fertility outcomes with limited definition of
male factor infertility or comparison with non-male factor infertility
cycles (4–9, 20). Ghazzawi et al. (21) did not find any difference in
fertilization rates (69%–79%) and pregnancy rates (PR; 21%–28%)
from use of ejaculated versus epididymal and testicular sperm in
ICSI cycles for severe oligoasthenoteratospermia, obstructive azoospermia,
and nonobstructive azoospermia. Live births per ET was
significantly lower with the nonobstructive group compared with
the other two groups (21). Van Steirteghem et al. (8) showed that
pregnancy outcomes from ejaculated, fresh and frozen-thawed epididymal
and testicular sperm were not significantly different with
34%–44% PRs. Forty-six percent of the patients using ejaculated
sperm had an abnormal semen analysis to define a male factor infertility
(8). Kamal et al. (22) recently reviewed 1,121 ICSI cycles for
obstructive azoospermia using epididymal sperm aspiration (percutaneous
or microscopic) versus testicular sperm (aspirated or biopsy).
No difference in fertilization, PR, or miscarriage rates was
found. These investigators did not have a nonobstructed male factor
infertility or a non-male factor infertility group for comparison. In
our analysis to study sperm acquisition techniques from a national
database we had to combine all cycles of ICSI that had a male factor
infertility component (i.e. mixed male plus female factor infertility
and male factor infertility only cycles) to generate enough numbers
for the comparison to non-male factor infertility cycles. The ICSI
cycles using ejaculated and aspirated sperm with male factor infertility
cycles had better CIG outcomes than non-male factor infertility
cycles. The LB rates were better with male factor infertility ejaculated
sperm only. Other types of sperm acquisition did not change outcomes
compared with non-male factor infertility cycles. Sperm acquisition
method provides the only information on the severity of male factor
infertility in the SART CORS. This information, however, does not indicate
the underlying cause for the method of sperm acquisition (e.g.,
obstructive [mainly vasectomy] and nonobstructive cause for biopsy).
Also, the type of testicular biopsy/degree of testicular sampling cannot
be ascertained (e.g., random open testicular sperm extraction
[TESE] for obstruction versus micro-TESE for nonobstructive azoospermia
or crytoozspermia). Aspiration was most likely performed
for an obstructive cause, but here again the type of aspiration is not
defined in the SART CORS (i.e., testis sperm aspiration [TESA] or
epididymis—percutaneous [PESA] or microsurgical epididymal
sperm aspiration [MESA]). Our analysis is the first to highlight
the limited information available on a national level but does not
end the debate regarding the efficacy of using different methods of
sperm acquisition.
The study has several limitations due the nature of the datasets.
The SART CORS does not indicate male age, which would be helpful
to determine how significant this is as a factor for ART outcomes.
There is a check box for ‘‘male factor’’ as the cause of the infertility
but this information is not subdivided into different male infertility
diagnoses as they exist for female factor infertility. The severity of
the male factor infertility cannot be determined. The definitions of
and reasons/diagnoses for the various sperm acquisition methods
is not stated. Outcome data may also be biased because more cycles
for male factor infertility involved ICSI compared with non-ICSI.
Pregnancy rate in relation to fertilization rate per cycle or number
of cycles per couple could not be determined due to some incomplete
reporting of the information to the SART-CORS. The CIG
and LB were believed to be more significant end points for issues
of fertilization and implantation, which can still be a concern from
male factor infertility (e.g., DNA fragmentation). We recognize
the burden that is already placed on ART centers to report a large
number of data points but an increased emphasis on male factor infertility
is recommended considering this contributes up to 50% of
infertility in the United States (1).
In conclusion, our study showed that ICSI for male factor infertility
had a worse outcome for CIG than non-male factor infertility
cycles and non-ICSI male factor infertility cycles. The LB rate
was not affected. A concern regarding the effect/severity of male
factor infertility in ART is suggested, but limited by the database.
The ICSI cycles with ejaculated and aspirated sperm for male factor
infertility had better pregnancy outcomes than non-male factor
infertility cycles. We suggest that SART consider adding additional
information on male factor infertility to the SART CORS database to
allow a more complete analysis of these male factor infertility
issues.
Acknowledgments: SART thanks all of its members for providing clinical information
to the SART CORS database for use by patients and researchers.
Without the efforts of SART members, this research would not have been
possible.
REFERENCES
1. World Health Organanization. Recent advances in
medically assisted conception. Report of a WHO Scientific
Group. World Health Organ Tech Rep Ser
1992;820:1–111.
2. Anderson JE, Farr SL, Jamieson DJ, Warner L,
Macaluso M. Infertility services reported by men in
the United States: national survey data. Fertil Steril
2009;91:2466–70.
3. Hashimoto H, Ishikawa T, Goto S, Kokeguchi S,
Fujisawa M, Shiotani M. The effects of severity of
oligozoospermia on intracytoplasmic sperm injection
(ICSI) cycle outcome. Syst Biol Reprod Med
2010;56:91–5.
4. Verza S, Esteves SC. Sperm defect severity rather
than sperm source is associated with lower fertilization
rates after intracytoplasmic sperm injection. Int
Braz J Urol 2008;34:49–56.
5. Vendrell JM, Aran B, Veiga A, Garcıa F, Coroleu B,
Egozcue S, et al. Spermatogenic patterns and early
embryo development after intracytoplasmic sperm
injection in severe oligoasthenozoospermia. J Assist
Reprod Genet 2003;20:106–12.
6. Ludwig M, Katalinic A. Pregnancy course and
health of children born after ICSI depending on parameters
of male factor infertility. Hum Reprod
2003;18:351–7.
7. Balaban B, Urman B, Isiklar A, Alatas C,
Mercan R, Aksoy S, et al. Blastocyst transfer following
intracytoplasmic injection of ejaculated, epididymal
or testicular spermatozoa. Hum Reprod
2001;16:125–9.
8. Van Steirteghem A, Nagy P, Joris H,
Janssenswillen C, Staessen C, Verheyen G, et al.
Results of intracytoplasmic sperm injection with
ejaculated, fresh and frozen-thawed epididymal and
testicular spermatozoa. Hum Reprod 1998;13(Suppl
1):134–42.
9. Nagy Z, Liu J, Cecile J, Silber S, Devroey P, Van
Steirteghem A. Using ejaculated, fresh, and frozen-
Fertility and Sterility â 613

thawed epididymal and testicular spermatozoa gives rise
to comparable results after intracytoplasmic sperm injection.
Fertil Steril 1995;63:808–15.
10. Meacham RB, Joyce GF, Wise M, Kparker A,
Niederberger C, Project UDiA. Male infertility.
J Urol 2007;177:2058–66.
11. Chandra A, Martinez GM, Mosher WD, Abma JC,
Jones J. Fertility, family planning, and reproductive
health of U.S. women: data from the 2002 National
Survey of Family Growth. Vital Health Stat
2005;23:1–160.
12. Centers for Disease Control and Prevention, American
Society of Reproductive Medicine, Society for Assisted
Reproductive Technology. 2008 Assisted reproductive
technology success rates: national summary and
fertility clinic reports. In: Atlanta: U.S. Department
of Health and Human Services, 2010.
13. States U. Fertility Clinic Success Rate and Certification
Act of 1992: Public Law 102-493. US Statut
Large 1992;106:3146–52.
14. Matzuk MM, Lamb DJ. The biology of infertility: research
advances and clinical challenges. Nat Med
2008;14:1197–213.
15. K€upker W, Schwinger E, Hiort O, Ludwig M,
Nikolettos N, Schlegel PN, et al. Genetics of male
subfertility: consequences for the clinical work-up.
Hum Reprod 1999;14(Suppl 1):24–37.
16. Mau Kai C, Juul A, McElreavey K, Ottesen AM,
Garn ID, Main KM, et al. Sons conceived by assisted
reproduction techniques inherit deletions in the azoospermia
factor (AZF) region of the Y chromosome
and the DAZ gene copy number. Hum Reprod
2008;23:1669–78.
17. Choi JM, Chung P, Veeck L, Mielnik A, Palermo GD,
Schlegel PN. AZF microdeletions of the Y chromosome
and in vitro fertilization outcome. Fertil Steril
2004;81:337–41.
18. Palermo GD, Neri QV, Takeuchi T, Squires J, Moy F,
Rosenwaks Z. Genetic and epigenetic characteristics of
ICSI children. Reprod Biomed Online 2008;17:820–33.
19. Katagiri Y, Neri QV, Takeuchi T, Schlegel PN,
Megid WA, Kent-First M, et al. Y chromosome assessment
and its implications for the development
of ICSI children. Reprod Biomed Online
2004;8:307–18.
20. Nicopoullos JD, Gilling-Smith C, Almeida PA,
Norman-Taylor J, Grace I, Ramsay JW. Use of surgical
sperm retrieval in azoospermic men: a meta-analysis.
Fertil Steril 2004;82:691–701.
21. Ghazzawi IM, Sarraf MG, Taher MR, Khalifa FA.
Comparison of the fertilizing capability of spermatozoa
from ejaculates, epididymal aspirates and testicular
biopsies using intracytoplasmic sperm injection.
Hum Reprod 1998;13:348–52.
22. Kamal A, Fahmy I, Mansour R, Serour G,
Aboulghar M, Ramos L, et al. Does the outcome of
ICSI in cases of obstructive azoospermia depend on
the origin of the retrieved spermatozoa or the cause
of obstruction? A comparative analysis. Fertil Steril
2010;94:2135–40.
614 Nangia et al. Male factor infertility and ART outcomes in the USA Vol. 96, No. 3, September 2011