FERTILITY AND STERILITY威

VOL. 81, NO. 5, MAY 2004

Copyright ©2004 American Society for Reproductive Medicine
Published by Elsevier Inc.
Printed on acid-free paper in U.S.A.

Sperm chromatin structure assay (SCSA姞)

parameters are related to fertilization,
blastocyst development, and ongoing
pregnancy in in vitro fertilization and
intracytoplasmic sperm injection cycles
Michael R. Virro, M.D.,a Kjersten L. Larson-Cook, Ph.D.,b and
Donald P. Evenson, Ph.D.b
Markham Fertility Center, Markham, Ontario, Canada

Objective: To determine the relationship between sperm chromatin structure assay (SCSA) parameters (DNA
fragmentation index [DFI] and high DNA stainability [HDS]), and conventional IVF and IVF/intracytoplas-
mic sperm injection (ICSI) outcomes.
Design: Retrospective review and prospective study.
Setting: Private IVF clinic.
Patient(s): Two hundred forty-nine couples undergoing first IVF and/or ICSI cycle.
Intervention(s): IVF, ICSI, blastocyst culture.
Main Outcome Measure(s): DFI, HDS, conventional semen parameters, IVF, ICSI.
Result(s): IVF and ICSI fertilization rates were not statistically different between high- and low-DFI groups.
More men with ⱖ15% HDS had lower (⬍25% and ⬍50%) IVF fertilization rates. High DNA stainability was
not related to ICSI fertilization rates. High DNA stainability did not affect blastocyst rates or pregnancy
outcomes. Men with ⱖ30% DFI were at risk for low blastocyst rates (⬍30%) and no ongoing pregnancies.
Men with ⱖ30% DFI had more male factors. World Health Organization thresholds were not predictive of
Received May 14, 2003;
revised and accepted
ongoing pregnancy.
September 15, 2003. Conclusion(s): The relationship between HDS and poor IVF fertilization rates provides preliminary evidence
Supported in part by the that ICSI may be indicated in men with ⱖ15% HDS. Men with high levels of DNA fragmentation (ⱖ30%
Environmental Protection DFI) were at greater risk for low blastocyst rates and failure to initiate an ongoing pregnancy. The SCSA
Agency, grant no. provides valuable prognostic information to physicians counseling couples before IVF and/or ICSI cycles.
R827019, and SCSA (Fertil Steril威 2004;81:1289⫺95. ©2004 by American Society for Reproductive Medicine.)
Diagnostics. This is South
Dakota Agricultural Key Words: Sperm chromatin structure assay, SCSA, DNA fragmentation, DFI, HDS, IVF, ICSI, blastocyst
Experiment Station rate, pregnancy, spontaneous abortion
Publication no. 3378 of the
journal series.
Reprint requests: Donald P. Male factor as the etiology of infertility has influencing ICSI embryo development, the low
Evenson, Ph.D., Olson been shown to have a significant, negative ef- incidence of blastocyst development likely
Biochemistry Labs, ASC
149, South Dakota State fect on the number of embryos that develop to stems in part from the relatively high incidence
University, Brookings, the blastocyst stage (1, 2). Therefore, sperm of fragmented DNA and chromosomal abnor-
South Dakota 57007 (FAX: quality should be scrutinized to determine the malities within the sperm population selected
605-692-9730;
E-mail: scsa@brookings.net). role it plays in poor blastocyst development for ICSI (4). These data indicate that irrepara-
a
Markham Fertility Centre, and/or the failure to initiate a term pregnancy. ble abnormalities in the paternal genome affect
Markham, Ontario, Canada. Male factor infertility correlates significantly blastocyst development even when the sperm is
b
Olson Biochemistry Labs, with blastocyst numbers (1, 2) and quality (3). injected directly into the oocyte via ICSI (1).
South Dakota State
University, Brookings, Furthermore, blastocyst formation favors The embryonic genome is activated on day
South Dakota.
those zygotes resulting from IVF over intracy- 3, and its transcriptional products supersede the
0015-0282/04/$30.00 toplasmic sperm injection (ICSI) (4, 5). Al- regulatory control provided by maternal mes-
doi:10.1016/j.fertnstert.2003.
09.063 though technical aspects may play a role in sages stored in the oocyte. Blastocyst culture

1289
maintains the embryo in vitro during and after activation of nuclear chromatin (DNA and protein) structure of these
the embryonic genome, allowing scientists to witness the immature sperm is abnormal, with a characteristically high
earliest expression of an “errant paternal genome” (6). Em- level of DNA stainability (%HDS). In a study of semen
bryo abnormalities seen in vitro can be more directly samples collected monthly for 8 consecutive months, the
related to male factors because the results can be assessed SCSA parameters were more repeatable than the conven-
without the interference of confounding female factors tional semen parameters of sperm count, motility, and mor-
(e.g., uterine and endocrine abnormalities) that may lead phology (21). Other studies have shown that the DFI can be
to embryo wastage or miscarriage post-transfer (7). In an excellent indicator of environmental pollution (22), smok-
vitro assessment of the embryos allows the development of ing (23, 24), and exposure to industrial toxicants (25, 26).
noninvasive selection criteria against abnormal, nonviable Additional factors that affect human SCSA data include
embryos and may lessen the chance that a child will be febrile illness (27), age (24), and cancer (28 –30).
conceived from sperm with an abnormal genome (6). Fur- In a Science article published in 1980, the SCSA showed
thermore, blastocyst culture aids scientific endeavors de- a clear difference in DNA quality between proven fertile and
signed to study the role that the paternal genome plays in infertile men (no conceptions over 1 year) and bulls (11). In
blastocyst development, implantation, and clinical pregnan- a subsequent comprehensive male factor fertility study of
cies. couples attempting to conceive without assisted reproductive
Gaining a better understanding of the impact of male techniques, the SCSA distinguished between highly fertile,
factors in abnormal blastocyst development is critically im- moderately fertile, and sub-/infertile men (12). In this study,
portant to clinical practice as only 20% of human embryos pregnancy rates were significantly lower when men had a
implant after transfer to the uterus (8). Transfer of embryos high percentage of immature sperm (ⱖ15% HDS). Further-
at the blastocyst stage provides the possibility of selecting more, no couple achieved a pregnancy when the DFI was
better-quality embryos with a higher implantation potential ⱖ30% during the month of conception. Similarly, three
(5). Even more critical may be evidence indicating that the other studies including a total of 146 patients reported no
most significant phenotypic manifestations of paternal ge- sustained pregnancies after IUI/IVF/ICSI when the DFI was
nome abnormalities occur during postimplantation develop- ⬎27% in the raw ejaculate (13, 17, 18, 31).
ment (9). Therefore, genetic abnormalities in the paternal To our knowledge, the research reported here is the first
genome may be a significant source of miscarriages, 50% of to compare SCSA parameters (DFI and HDS) to IVF/ICSI
which are currently unexplained (7). fertilization, blastocyst development, spontaneous abortions,
The role of paternal genome abnormalities in miscarriage and chemical and ongoing pregnancies.
is underestimated because the assays used in clinical practice
measure major numerical or structural abnormalities but do MATERIALS AND METHODS
not measure minor but potentially global chromosomal dam-
age (DNA nicks, double strand breaks; 9). Sperm from Patients
infertile men have an elevated rate of these “minor” DNA Two hundred forty-nine couples commencing their first
abnormalities (10 –17), which may lead to the observed IVF cycle had SCSA testing done on the male partner. Data
abnormal blastocyst development, failed implantation, and from the first 63 couples were looked at retrospectively.
spontaneous miscarriages in conceptions using these sperm. After this, every couple planning a cycle had SCSA testing
If evaluated in men before IVF and/or ICSI, sperm DNA done before their egg retrieval. South Dakota State Univer-
abnormalities would likely identify the cause of infertility in sity’s Institutional Review Board approved this research
a large percentage of patients (18). Therefore, evaluating the study. Approval was not sought in the private IVF clinic
paternal genome may be a significant upgrade to the diag- because of the lack of such a review board. Patients were
nostic and prognostic information provided by the conven- grouped according to their SCSA data. Men with ⬍30% DFI
tional male factor infertility examination. were classified as having low levels of DNA fragmentation,
and men with ⱖ30% DFI were classified as having high
Over the past decade, several techniques have emerged as
levels of DNA fragmentation. A second, independent group-
potential protocols for evaluating chromatin/DNA integrity.
ing was based on HDS. Men with ⬍15% HDS were said to
The sperm chromatin structure assay (SCSA) measures the
have low levels of immature sperm, and men with ⱖ15%
percentage of sperm with a high susceptibility to low pH-
HDS were said to have high levels of immature sperm.
induced DNA denaturation and is expressed as the DNA
fragmentation index (%DFI). DNA fragmentation index is a SCSA Protocols
highly accurate, repeatable measure of DNA quality that is The SCSA protocol has been described elsewhere by
proportionate to the level of DNA strand breaks in sperm Evenson et al. (12, 32). Each semen sample was frozen in
(19, 20). In addition to quantifying DNA strand breaks LN2 without cryoprotectant and transported in a LN2 dry
through DFI, the SCSA simultaneously identifies the per- shipper via Federal Express to SCSA Diagnostics (Brook-
centage of sperm with immature nuclear development. The ings, SD) for SCSA testing. From each semen sample, two

1290 Virro et al. Sperm DNA predicts lower pregnancy rate Vol. 81, No. 5, May 2004
 FIGURE 1

Native (green fluorescence) versus fragmented DNA (red fluorescence) cytograms showing the two SCSA parameters of
interest. Left: semen sample with a high percentage of sperm with high levels of DNA fragmentation (%DFI). Right: semen
sample with a high percentage of sperm with HDS (%HDS).

Virro. Sperm DNA predicts lower pregnancy rate. Fertil Steril 2004.

separate measurements of 5,000 sperm cells each were eval- Seventy-five percent of patients had both conventional
uated by flow cytometry. High DNA stainability was calcu- IVF and ICSI performed, splitting the number of eggs
lated based on the percentage of sperm with high levels of equally for both procedures. Eighteen percent had all of their
green fluorescence (y-axis, Fig. 1). Native (green fluores- eggs injected for male factor reasons, and 7% had only
cence, y-axis) versus fragmented DNA (red fluorescence, conventional IVF performed. After egg retrieval, oocytes
x-axis) cytograms (Fig. 1) were used to determine the DFI of were rinsed of follicular fluid in HTF-Hepes ⫹ 10% HSA
the entire sperm population. The percentage of sperm with (human serum albumin; Bayer Corp. Canada, Toronto, On-
high levels of DNA fragmentation (%DFI) was quantified tario). Oocytes were transferred to organ culture dishes (Fal-
based on the increased ratio of red/[red ⫹ green] fluores- con, BD Biosciences, Franklin Lakes, NJ) that contained 800
cence. ␮L of pre-equilibrated P1 ⫹ 10% substitute serum supple-
ment (SSS; Irvine Scientific Canada, Mississauga, Ontario).
Assisted Reproductive Techniques Oocytes were incubated for 4 – 6 hours (37°C; 5% CO2, 5%
Stimulation Protocol and Oocyte Retrieval O2, and 90% N2) before insemination (IVF) or injection
(ICSI).
All women were suppressed with 0.1 cc of Lupron SC
(Abbott Laboratories Ltd., Montreal, Quebec, Canada) start- Fertilization Assessment, Embryo Culture, and Grading
ing on day 21. Suppression ranged from 13 to 20 days for Oocytes were examined for evidence of fertilization
IVF scheduling purposes. When E2 levels were below 100 16 –18 hours after insemination or injection. Normally fer-
pmol/L, ovarian stimulation was started with pure FSH [ei- tilized ova (2PN; 2PB) were transferred to organ culture
ther Gonal-F (Serono Canada Inc., Oakville, Ontario) or dishes containing 90 ␮L of pre-equilibrated P1 ⫹ 10% SSS.
Puregon (Organon Canada Ltd., Scarburough, Ontario)] in Zygotes were incubated for 48 hours at 37°C in 5% CO2, 5%
96% of patients. The remaining 4% of patients had either O2, and 90% N2. On day 3, embryos were transferred to
pure LH (Lhadi, Serono Canada, Ltd.) or Humegon (Or- organ culture dishes containing 90 ␮L of pre-equilibrated
ganon Canada Ltd.) added to their protocol. The mean num- blastocyst medium ⫹ 10% SSS (Irvine). Embryos were
ber of 75 unit ampoules injected was 26.5 per cycle, and the incubated for an additional 48 hours until day 5.
mean E2 level was 4,537 pmol/L at the time of hCG [Profasi
(Serono Canada Inc.) 10,000 units] administration. At least Blastocyst Grading and Transfer
two lead follicles were 18 mm at the time of hCG adminis- On the morning of day 5, embryos were examined for
tration as determined by transvaginal ultrasound. Egg re- further cleavage to the blastocyst stage. Blastocyst develop-
trieval was performed 35 hours after hCG injection. Egg ment was characterized according to Gardner and School-
retrieval was performed under neurolept anesthesia using craft’s blastocyst grading system (33). Any blastocyst that
Demerol, Propofol (Abbott Laboratories Ltd.), and Versed. scored less than 3 and that had more than 60 cells was graded

FERTILITY & STERILITY威 1291

 TABLE 1

Female and male factors within DNA fragmentation index (DFI) and high DNA stainability (HDS) groups.

Conventional semen parameters

Years of Female No. of mature Volume Count Motility Morphology

infertility age eggs (mL) (million) (% motile) (WHO)

DFI
Group 1, DFI ⬍30%, n ⫽ 178 4.7 ⫾ 3.0 34.1 ⫾ 3.8 8.7 ⫾ 4.1 3.2 ⫾ 1.4 107.6 ⫾ 90.2 49.1 ⫾ 17.0 53.4 ⫾ 16.5
Group 2, DFI ⱖ30%, n ⫽ 71 4.2 ⫾ 2.6 33.9 ⫾ 4.4 8.5 ⫾ 4.1 3.4 ⫾ 1.3 54.1a ⫾ 66.0 35.2b ⫾ 15.6 41.0c ⫾ 17.0
HDS
Group 1, HDS ⬍15%, n ⫽ 207 4.5 ⫾ 2.9 34.2 ⫾ 4.0 8.5 ⫾ 4.1 3.2 ⫾ 1.3 104.6 ⫾ 88.5 47.5 ⫾ 17.2 51.7 ⫾ 16.7
Group 2, HDS ⱖ15%, n ⫽ 42 4.4 ⫾ 2.7 33.1 ⫾ 3.8 9.1 ⫾ 4.1 3.6 ⫾ 1.7 32.3b ⫾ 47.4 33.8b ⫾ 16.4 40.6a ⫾ 18.9

Note: Values represent mean ⫾ SD for DFI and HDS groups. Values within column and within sperm chromatin structure assay variable (DFI or HDS) with
different superscripts are significantly different. WHO ⫽ World Health Organization.
a
P⬍.01.
b
P⬍.0001.
c
P⬍.001.
Virro. Sperm DNA predicts lower pregnancy rate. Fertil Steril 2004.

as an early blastocyst. Immediately before transfer, the blas- meaningful linear relationships (regression analysis) were
tocysts were transferred to HTF-Hepes ⫹ 30% SSS. Blas- identified between IVF/ICSI outcomes and conventional se-
tocysts were loaded in 10 –20 ␮L of HTF-Hepes ⫹ 30% SSS men parameters (volume, count, motility, or morphology) or
in the tip of an Edwards-Wallace embryo replacement cath- SCSA parameters (DFI or HDS). Analysis of variance and
eter (Cooper Surgical, Shelton, CT). Only day 5 blastocyst regression analysis were used to determine the relationships
data were used in this manuscript. All embryos were graded between SCSA parameters and conventional semen param-
by one embryologist who was blinded to the %DFI and eters as well as female age, years of infertility, and total
%HDS for the sperm sample used in that couple’s cycle. number of mature eggs.
Couples had up to but not greater than two blastocysts
transferred on the fifth day after egg retrieval. Embryo trans- RESULTS
fer (ET) was performed under ultrasound guidance using an
Edwards-Wallace catheter (Cooper Surgical). Patients were SCSA Groups
requested to remain supine for 30 minutes after transfer and The 249 IVF and/or ICSI patients included in the study
then have limited activity for the next 48 hours. Serum were divided based on their SCSA-defined DNA fragmen-
␤hCG levels were drawn 12 days after ET. All patients were tation levels, low %DFI (⬍30% DFI, n ⫽ 178) and high
supported with P vaginal suppositories (200 mg) 24 hours %DFI (ⱖ30% DFI, n ⫽ 71). Men were also divided based on
after egg retrieval. If the ␤hCG level was greater than 200 on their SCSA-defined DNA stainability, low HDS (⬍15%
day 12, patients started hCG support (2,500 units) IM every HDS, n ⫽ 207) and high HDS (ⱖ15% HDS, n ⫽ 42).
3 days until the twelfth week of gestation. Ultrasound was Female age, years of infertility, and average number of
performed between 6 and 6.5 weeks to confirm fetal viabil- mature eggs collected from the female partners did not vary
ity. Women with a positive ␤hCG without a pregnancy at the significantly between the DFI or HDS groups (Table 1).
time of ultrasound or that miscarried before the twelfth week
were identified as having a spontaneous abortion. Conventional Semen Parameters and DFI
Men with ⱖ30% DFI were 6.9 times (3.8 –12.62, confi-
Statistical Analysis dence bounds) more likely to have one or more abnormal
Each semen sample was measured twice in tandem by the conventional semen parameter(s) and therefore be identified
SCSA. Statistical analysis was completed using the mean as “male factor.” The means of sperm count (P⫽.001),
DFI and HDS of the replicate runs. Pearson’s ␹2 test was motility (P⬍.0001), and World Health Organization (WHO)
used to measure the relationship between DFI groups morphology (P⫽.0002) were significantly lower in men with
(ⱖ30% DFI vs. ⬍30% DFI) and HDS groups (ⱖ15% HDS ⱖ30% DFI (Table 1). Individually, motility (r2 ⫽ 0.17,
vs. ⬍15% HDS) and the presence of male factors and P⬍.0001), count (r2 ⫽ 0.12, P⬍.0001), and WHO morphol-
IVF/ICSI outcomes [low fertilization rate (⬍25% and ogy (r2 ⫽ 0.13, P⬍.0001) were significant predictors of DFI.
⬍50%), low blastocyst rate (⬍30%), spontaneous abortion, In multiple regression analysis, motility and count signifi-
and chemical and ongoing pregnancy]. No biologically cantly predicted 21% (P⬍.01) of the variation in DFI.

1292 Virro et al. Sperm DNA predicts lower pregnancy rate Vol. 81, No. 5, May 2004
 TABLE 2

Percentage of couples with low fertilization rates (⬍25% and ⬍50%) in DNA fragmentation index (DFI) and high DNA
stainability (HDS) groups.

Fertilization rate ⬍25% Fertilization rate ⬍50%

IVF and IVF and

IVF ICSI ICSI IVF ICSI ICSI

DFI
Group 1, DFI ⬍30% 9 4 2 28 12 18
(n ⫽ 13/144) (n ⫽ 4/114) (n ⫽ 4/178) (n ⫽ 40/144) (n ⫽ 14/114) (n ⫽ 32/178)
Group 2, DFI ⱖ30% 24 7 7 32 23 21
(n ⫽ 8/34) (n ⫽ 4/60) (n ⫽ 5/70) (n ⫽ 11/34) (n ⫽ 14/60) (n ⫽ 15/70)
HDS
Group 1, HDS ⬍15% 8 5 3 25 16 17
(n ⫽ 13/157) (n ⫽ 7/138) (n ⫽ 6/206) (n ⫽ 40/157) (n ⫽ 22/138) (n ⫽ 36/206)
Group 2, HDS ⱖ15% 38a 3 7 52b 17 26
(n ⫽ 8/21) (n ⫽ 1/36) (n ⫽ 3/42) (n ⫽ 11/21) (n ⫽ 6/36) (n ⫽ 11/42)
Note: Values within a column and within sperm chromatin structure assay variable (DFI or HDS) with different superscripts are significantly different. Values
are percents.
a
P⬍.001.
b
P⬍.01.
Virro. Sperm DNA predicts lower pregnancy rate. Fertil Steril 2004.

Conventional Semen Parameters and HDS men with ⱖ30% DFI. There was no significant relationship
Men with ⱖ15% HDS were 4.8 times (2.4 –9.3, confi- between ⱖ15% HDS and poor blastocyst rates. Furthermore,
dence bounds) more likely to have one or more abnormal no WHO threshold for semen volume, sperm count, motility,
conventional semen parameter(s) and therefore be identified or morphology was predictive of poor blastocyst rates.
as male factor. The means of sperm count (P⬍.0001), mo- Pregnancy Outcomes
tility (P⬍.0001), and WHO morphology (P⫽.004) were Men with high levels of DNA fragmentation (ⱖ30% DFI)
significantly lower in men with ⱖ15% HDS (Table 1). had a lower chance of initiating a chemical pregnancy than
Individually, motility (r2 ⫽ 0.09, P⬍.0001), count (r2 ⫽
0.11, P⬍.0001), and WHO morphology (r2 ⫽ 0.09,
P⬍.0001) were significant predictors of HDS. In multiple
regression analysis, motility and count significantly pre- FIGURE 2
dicted 15% (P⬍.01) of the variation in HDS.
Percentage of couples with a low blastocyst rate (⬍30%)
Fertilization Rates and SCSA Parameters between DFI groups. No significant differences were found
between HDS groups and blastocyst rate.
Although trends were evident, fertilization rates were not
statistically different between the high- and low-DFI groups
(Table 2). However, the percentage of couples with low
(⬍25%, P⫽.001) and moderately low (⬍50%, P⫽.01) IVF
fertilization rates was significantly higher in couples includ-
ing men with ⱖ15% HDS (Table 2). There was no relation-
ship between ICSI fertilization rates and HDS. After ICSI,
men with ⬎15% HDS were at an equal risk (0.392–2.8,
confidence bounds) of having ⬍50% fertilization rates as
those men with ⬍15% HDS. Combined, IVF and ICSI
fertilization rates were not significantly related to HDS.

Blastocyst Development and SCSA

Parameters
Overall, men with ⱖ30% DFI had a greater chance of
having poor (⬍30%) blastocyst rates (P⫽.003; Fig. 2).
When IVF and ICSI data were divided, there was still a trend
for low blastocyst rates in IVF (P⫽.07) and ICSI (P⫽.03) in Virro. Sperm DNA predicts lower pregnancy rate. Fertil Steril 2004.

FERTILITY & STERILITY威 1293

 men (⬍30% DFI) that had a 47% chance of initiating a term
FIGURE 3 pregnancy.
Female partners of men within group 2 (ⱖ30% DFI) had a Ahmadi and Ng (35) have previously reported in a mouse
lower rate of chemical pregnancy (P⫽.02) and a higher rate of model system that spermatozoa with defective DNA can
spontaneous abortion (P⫽.11) and a significant decrease in
pregnancies ongoing at 12 weeks of gestation (P⬍.01). HDS fertilize an oocyte and produce high-quality early-stage em-
groups and the WHO thresholds for normal semen analysis bryos, but then, as the extent of the DNA damage increases,
(volume, density, motility, and morphology) were not signifi- the likelihood of a successful term pregnancy decreases.
cantly related to these measures of pregnancy outcome.
Similarly, this and previous studies show that high levels of
DNA fragmentation were not significantly associated with a
decrease in fertilization rates (13, 14, 17, 34, 36). Instead, the
deleterious consequences of fragmented paternal DNA be-
came evident when the embryonic genome was activated.
Specifically, high levels of sperm DNA fragmentation
(ⱖ30% DFI) were manifest in a significant decrease in
blastocyst and ongoing pregnancy rates with a trend toward
a lower rate of chemical pregnancies and a higher rate of
spontaneous abortions.
The relationship between a high percentage of immature
sperm (ⱖ15% HDS) and low IVF fertilization rates supports
the finding from an in vivo fertility study that showed that
HDS was related to a decrease in pregnancies (12). Of
clinical significance, ICSI fertilization rates were not related
to HDS. Therefore, ICSI apparently compensated for the
Virro. Sperm DNA predicts lower pregnancy rate. Fertil Steril 2004.
sperm functional abnormalities associated with HDS and
eliminated any significant influence that a high percentage of
immature sperm had on fertilization rates. It is possible that
ICSI is bypassing a natural selection barrier, which ensures
men with ⬍30% DFI (P⫽.02; Fig. 3). This trend was com-
that fertilization will not occur when the chromatin is not
pounded by the increase in spontaneous abortions in men
properly organized. Yet immature sperm (HDS population)
with ⱖ30% DFI (P⫽.11; Fig. 3). The cumulative effect of
do not have an increase in the level of DNA fragmentation
these trends lead to the significant decrease in pregnancy
(DFI population). Therefore, the negative effect that imma-
rates found in men with high levels of DNA (28%) fragmen-
ture sperm has on pregnancy rates is likely due to intact
tation versus those men with ⬍30% DFI (47%, P⬍.01; Fig.
DNA being complexed with precursor protamines (23) or an
3). There was no relationship between HDS and chemical
altered ratio of protamine forms (37).
pregnancies, spontaneous abortions, or ongoing pregnancies.
Similarly, no WHO thresholds for normal conventional se- Depending on the couple’s history and number of repro-
men parameters were significantly predictive of pregnancy ductive treatment cycles, the SCSA test may give couples a
outcomes. long-awaited answer to the cause of their infertility. Sperm
donation may be a viable alternative for some of these
couples, especially when poor blastocyst quality and re-
DISCUSSION peated failed cycles are thought to be secondary to sperm
Most fertility clinics evaluate semen samples simply by DNA fragmentation. In our program, 12 couples elected for
conventional analysis, which does not ensure the absence of treatment with donor sperm after discussion about their
a male factor problem (31, 34). Although men with abnormal abnormal SCSA results (ⱖ30% DFI). These 12 couples had
semen parameters were at a significantly greater risk of experienced poor blastocyst development (no high-grade
having high levels of DNA fragmentation, the best multiple blastocysts available for transfer) and unsuccessful preg-
regression model, including count and motility, predicted nancy outcomes. Nine of these 12 couples conceived within
only 21% and 15% percent of the variation in DFI and HDS, three cycles of donor insemination, ending years of infertil-
respectively. When all conventional semen parameters were ity. The positive results with donor sperm indicate that the
normal, 18% of the men still had ⱖ30% DFI and fell into the damaged paternal genome of the male partners’ sperm con-
high-risk category for poor blastocyst development and fail- tributed to poor preimplantation embryo development in the
ure to initiate an ongoing pregnancy. Equally important, couples’ previous cycles. These results are supported by
39% of the men with abnormal semen parameters did not other studies showing that poor embryo quality (2), de-
have high levels of DNA fragmentation. These men were creased implantation rates, and increased spontaneous abor-
effectively treated with IVF/ICSI and fell into the group of tion rates associated with severe male factors (38) improve

1294 Virro et al. Sperm DNA predicts lower pregnancy rate Vol. 81, No. 5, May 2004
when the women were inseminated with donor sperm (39, and its relationship to fertilization and embryo development. Hum
Reprod 2002;17:990 –8.
40). Therefore, insight into the DNA integrity of the sperm 15. Zini A, Bielecki R, Phang D, Zenzes MT. Correlations between two
offer a direct explanation for poor embryo development and markers of DNA integrity and, DNA denaturation and DNA fragmen-
tation, in fertile and infertile men. Fertil Steril 2001;75:674 –7.
pregnancy rates in our patients and lead to more successful 16. Duran EH, Morshedi M, Taylor S, Oehninger S. Sperm DNA quality
treatments. predicts intrauterine insemination outcome: a prospective cohort study.
Hum Reprod 2002;17:3122–8.
The SCSA test is a valuable noninvasive assay with 17. Larson-Cook KL, Brannian JD, Hansen KA, Kasperson KM, Aamold
ET, Evenson DP. Relationship between the outcomes of assisted repro-
adequate sensitivity and repeatability to identify men at ductive techniques and sperm DNA fragmentation as measured by the
increased risk of poor blastocyst rates and negative preg- sperm chromatin structure assay. Fertil Steril 2003;80:895–902.
18. Agarwal A, Said TM. Role of sperm chromatin abnormalities and DNA
nancy outcomes in a clinical setting. Men with ⱖ30% DFI damage in male infertility. Hum Reprod Update 2003;9:331–45.
are at an increased risk of having poor blastocyst develop- 19. Aravindan GR, Bjordahl J, Jost LK, Evenson DP. Susceptibility of
human sperm to in situ DNA denaturation is strongly correlated with
ment even though their fertilization rates are not significantly DNA strand breaks identified by single-cell electrophoresis. Exp Cell
lower than men with ⬍30% DFI. Therefore, culturing zy- Res 1997;236:231–7.
20. Sailer BL, Jost LK, Evenson DP. Mammalian sperm DNA susceptibil-
gotes to the blastocyst stage may allow the selection of ity to in situ denaturation associated with the presence of DNA strand
blastocysts after the activation of the embryonic genome and breaks as measured by the terminal deoxynucleotidyl transferase assay
(TDTA). J Androl 1995;16:80 –7.
offer these couples an increased likelihood of pregnancy. 21. Evenson DP, Jost LK, Baer RK, Turner TW, Schrader SM. Individu-
ality of DNA denaturation patterns in human sperm as measured by the
Performing the SCSA test on a semen sample before sperm chromatin structure assay. Reprod Tox 1991;5:115–25.
undergoing IVF or ICSI can offer couples insight into their 22. Selevan SG, Borovec L, Slott VL, Zudov Z, Rube, Evenson DP, et al.
Semen quality and reproductive health of young Czech men exposed to
risk of these negative outcomes that is not provided by seasonal air pollution. Environ Health Perspec 2000;108:887–94.
standard semen analysis alone. Physicians may advise cou- 23. Potts RJ, Newbury CJ, Smith G, Notarianni LJ, Jefferies TM. Sperm
chromatin changes associated with male smoking. Mutat Res 1999;423:
ples that have ⱖ30% DFI that their SCSA test score indi- 103–11.
cates they have a significantly lower chance of achieving an 24. Spano M, Kolstad AH, Larson SB, Cordelli E, Leter G, Giwercman A,
et al. The applicability of the flow cytometric sperm chromatin structure
ongoing pregnancy due to the quality of the paternal ge- assay in epidemiological studies. Hum Reprod 1998:2495–505.
nome. At this juncture, we believe that the SCSA makes a 25. Grajewski B, Cox C, Schrader SM, Murray WE, Edwards RM, Turner
T, et al. Semen quality and hormone levels among radiofrequency heat
solid contribution to the semen analysis profile and may be sealer operators. J Occup Env Med 2000;42:993–1005.
an effective diagnostic and prognostic tool to evaluate male 26. Lemasters GK, Olsen DM, Yiin JH, Lockey JE, Shukla R, Selevan SG,
et al. Male reproductive effects of solvent and fuel exposure during
factor infertility. aircraft maintenance. Reprod Toxic 1999;13:155–66.
27. Evenson DP, Jost LK, Corzett M, Balhorn R. Effect of elevated body
References temperature on human sperm chromatin structure. J Androl 2000;21:
1. Jones GM, Trounson AO, Lolatgis N, Wood C. Factors affecting the 739 –46.
success of human blastocyst development and pregnancy following in 28. Kobayashi H, Larson K, Sharma R, Nelson DR, Evenson DP, Toma H,
vitro fertilization and embryo transfer. Fertil Steril 1998;70:1022–9. et al. DNA damage in patients with untreated cancer as measured by the
2. Janny L, Menezo YJR. Evidence for a strong paternal effect on human sperm chromatin structure assay. Fertil Steril 2001;75:469 –75.
preimplantation embryo development and blastocyst formation. Mol 29. Fossa SC, De Angelis P, Draggerud SM, Evenson D, Theodorsen L,
Reprod Dev 1994;38:36 –42. Claussen OP. Prediction of posttreatment spermatogenesis in patients
3. Miller J, Smith T. The effect of intracytoplasmic sperm injection and with testicular cancer by flow cytometric sperm chromatin structure
semen parameters on blastocyst development in vitro. Hum Reprod assay. Cytometry 1997;30:192–6.
2001;16:918 –24. 30. Evenson DP, Klein FA, Whitmore WF, Melamed MR. Flow cytometric
4. Dumoulin JCM, Coonen E, Bras M, Bergers-Janssen JM, Ignoul- evaluation of sperm from patients with testicular carcinoma. J Urol
Vanvuchelen RCM, van Wissen LCP, et al. Embryo development and 1984;132:1220 –5.
chromosomal anomalies after ICSI: effect of the injection procedure. 31. Saleh RA, Agarwal A, Nelson DR, Nada EA, El-Tonsy MH, Alvarez
Hum Reprod 2001;16:306 –12. JG, et al. Increased sperm nuclear DNA damage in normozoospermic
5. Shoukir Y, Chardonnens D, Campana A, Sakkas D. Blastocyst devel- infertile men: a prospective study. Fertil Steril 2002;78:313–8.
opment from supernumerary embryos after intracytoplasmic sperm 32. Evenson DP, Larson KL, Jost LK. Sperm chromatin structure assay: its
injection: a paternal influence. Hum Reprod 1998;13:1632–7. clinical use for detecting sperm DNA fragmentation in male infertility
6. Sakkas D. The use of blastocyst culture to avoid inheritance of an and comparisons with other techniques. J Androl 2002;23:25–43.
abnormal paternal genome after ICSI. Hum Reprod 1999;14:4 –5. 33. Gardner DK, Schoolcraft WB. In-vitro culture of human blastocysts. In:
7. Li TC. Guides for practitioners. Recurrent miscarriage: principles of Jansen R, Mortimer D, eds. Towards reproductive certainty: infertility
management. Hum Reprod 1998;13:478 –82. and genetics beyond 1999. Carnforth: Parthenon Press, 1999:378 –88.
8. Edwards RG, Beard HK. Blastocyst stage transfer: pitfalls and benefits. 34. Sakkas D, Urner F, Bianchi PG, Bizzaro D, Wagner I, Jaquenoud N, et
Hum Reprod 1999;14:1–6. al. Sperm chromatin abnormalities can influence decondensation after
9. Banerjee S, Lamond S, McMahon A, Campbell S, Nargund G. Does intracytoplasmic sperm injection. Hum Reprod 1996;11:837–43.
blastocyst culture eliminate paternal chromosomal defects and select 35. Ahmadi A, Ng SC. Fertilizing ability of DNA-damaged spermatozoa. J
good embryos? Inheritance of an abnormal paternal genome following Exp Zool 1999;284:696 –704.
ICSI. Hum Reprod 2000;15:2455–9. 36. Twigg JP, Irvine DS, Aitken RJ. Oxidative damage to DNA in human
10. Irvine DS, Twigg JP, Gordan EL, Fulton N, Milne A, Aitken J. DNA sperm does not preclude pronucleus formation at intracytoplasmic
integrity in human spermatozoa: relationships with semen quality. J sperm injection. Hum Reprod 1998;13:1864 –71.
Androl 2000;21:33–44. 37. de Yebra L, Ballescá JL, Vanrell JA, Corzett M, Balhorn R, Oliva R.
11. Evenson DP, Darvynkiewicz Z, Melamed MR. Relation of mammalian Detection of P2 precursors in the sperm cells of infertile patients who
sperm chromatin heterogeneity to fertility. Science 1980;240:1131–3. have reduced protamine P2 levels. Fertil Steril 1998;69:755–9.
12. Evenson DP, Jost LK, Marshall D, Zinaman MJ, Clegg E, Purvis K, et 38. Keifer D, Check JH, Katsoff D. Evidence that oligoasthenozoospermia
al. Utility of sperm chromatin structure assay as a diagnostic tool in the may be an etiologic factor for spontaneous abortions after in vitro
human fertility clinic. Hum Reprod 1999;14:1039 –49. fertilization– embryo transfer. Fertil Steril 1997;68:545–8.
13. Larson KL, De Jonge CJ, Barnes AM, Jost LK, Evenson DP. Relation- 39. Genesca A, Caballin MR, Miro R, Benet J, Germá JR, Egozcue J.
ship of assisted reproductive technique (ART) outcomes with sperm Repair of human sperm chromosome aberrations in the hamster egg.
chromatin integrity and maturity as measured by the sperm chromatin Hum Genet 1992;82:1816.
structure assay (SCSA). Hum Reprod 2000;15:1717–22. 40. Obasaju M, Kadam A, Sultan K, Fateh M, Munné S. Sperm quality may
14. Morris ID, Ilott S, Dixon L, Brison DR. The spectrum of DNA damage adversely affect the chromosome constitution of embryos that result
in human sperm assessed by single cell electrophoresis (COMET assay) from intracytoplasmic sperm injection. Fertil Steril 1999;72:1113–25.

FERTILITY & STERILITY威 1295