In vitro fertilization–intracytoplasmicsperm injection outcome in patientswith a markedly high DNAfragmentation index (>50%)
Shir Dar, M.D.,a,b,c Stephanie A. Grover, M.Sc.,a Sergey I. Moskovtsev, M.D., Ph.D.,a,b Sonja Swanson, M.Sc.,a
Ari Baratz, M.D.,a,b,c and Clifford L. Librach, M.D.a,b,c
a CReATe Fertility Center, b Department of Obstetrics andGynecology, University of Toronto, and c Division of ReproductiveEndocrinology and Infertility, Department of Obstetrics and Gynecology, Sunnybrook Health Sciences Centre andWomen'sCollege Hospital, Toronto, Ontario, Canada
Objective: To investigate differences in fertilization, clinical pregnancy, and miscarriage rates between men with a markedly highsperm DNA fragmentation index (DFI) (>50%) and those with low DFI (%15%) in couples matched by female partner age and ovarianreserve as determined by antim€ullerian hormone (AMH) level.Design: Retrospective cohort study.Setting: University-affiliated fertility center.Patient(s): Men undergoing intracytoplasmic sperm injection (ICSI) cycles who had low (n ¼ 114) or markedly high (n ¼ 36) DNAdamage.Intervention(s): None.Main Outcome Measure(s): Sperm DNA damage evaluated by acridine orange flow cytometry and expressed as the DFI, with thepotential confounders of ovarian reserve and age controlled for by multivariable logistic regression analysis.Result(s): The fertilization and clinical pregnancy rates were not different between the two groups. We observed a trend towarda higher miscarriage rate with the high DFI group, but it did not reach statistical significance.Conclusion(s): Intracytoplasmic sperm injection in men with a high DFI with sperm selected by movement and morphologycharacteristics resulted in a similar pregnancy rate compared with the controls with a normal DFI. However, the trend observed of
Use your smartphone
an increase in miscarriages suggests that any potential negative impact may appear later indevelopment. Future studies involving a larger cohort may determine if the miscarriage trendreaches statistical significance. (Fertil Steril� 2013;100:75–80. �2013 by American Societyfor Reproductive Medicine.)Key Words: DNA Fragmentation Index, ICSI, spermatozoa
Discuss: You can discuss this article with its authors and with other ASRM members at http://fertstertforum.com/dars-ivf-icsi-dna-fragmentation-index-spermatozoa/
to scan this QR codeand connect to thediscussion forum forthis article now.*
* Download a free QR code scanner by searching for “QRscanner” in your smartphone’s app store or app marketplace.
pproximately one of every six motility, and morphology. This type of has been increasingly used as an
A couples will experience infer-tility, and about 40% of these
cases are male factor infertility (1, 2).The ‘‘standard’’ semen analysis usingphase contrast microscopy primarilyinvolves assessing four parameters:semen volume, sperm concentration,
ReceivedNovember 8, 2012; revised February 25, 2013, 2013.
S.D. has nothing to disclose. S.A.G. has nothing tonothing to disclose. A.B. has nothing to disclose
analysis provides clinicians with thebasic information they need to maketheir initial assessment of the malepartner. However, these parametersalone have relatively low predictivevalue for fertility and sperm functionassessment (3). Sperm DNA testing
3; acceptedMarch 1, 2013; published online April
disclose. S.I.M. has nothing to disclose. S.S. has. C.L.L. has nothing to disclose.ty Centre, 790 Bay Street, Suite 1100, Toronto,mail.com).
5-0282/$36.00Medicine, Published by Elsevier Inc.
adjunct to the standard spermparameters (4–7). As a biomarker forfertility, numerous studies have shownthat high sperm DNA damage isassociated with an increased time toconception, impaired embryocleavage, higher miscarriage rates,and recurrent pregnancy loss afterin vitro fertilization (IVF) andintracytoplasmic sperm injection(ICSI) (6–10). When sperm withoxidative DNA damage is used forintrauterine insemination (IUI), IVF, orICSI, it has been found that the sperm
still have the potential to fertilize an oocyte and thereby resultin a pregnancy. However, if damaged sperm DNA isintegrated into the embryonic genome, it may lead to errorsin DNA replication, transcription, and translation duringembryonic development, contributing to a number ofhuman diseases or fetal loss (11, 12). Hence, sperm DNAdamage may affect the health of some children conceivedusing assisted reproductive technologies (ART) both in theshort and long-term.
A higher incidence of malformations and diseases havebeen found in children of subfertile couples (>12 months toconception) that conceived either by assisted reproductiontechnology (ART) or spontaneously. Because high DNA dam-age is more frequent in these couples, damaged DNA may bea contributing factor for the abnormalities in some of thesecases (13). Childhood cancers such as leukemia as well asautism have been associated with poor sperm DNA integrity,and these effects were found to be enhanced by paternalsmoking (4, 14, 15). The aim of our study was to look at thefertilization, pregnancy, and miscarriage rates in patientswith markedly high DNA fragmentation index (DFI >50%)as compared with a group of men with normal DFI whounderwent an ICSI cycle. The groups were stratified by ageand the female partner's ovarian reserve as measured byantim€ullerian hormone (AMH) level.
MATERIALS AND METHODSA retrospective cohort study was performed that reviewed theclinical charts of 97 patients who underwent 152 IVF-ICSIcycles at a university-affiliated fertility center (the CReATeFertility Centre). The study was approved by the institutionalresearch ethics board. The inclusion criteria were couples whohad an IVF-ICSI cycle from 2007–2012 where the malepartner either had a markedly high (>50%) or low (<15%)sperm DFI. Group 1 (n ¼ 39 cycles) included all the cyclesof patients who had a high DFI; group 2 (controls, n ¼ 115cycles) were consecutive cases of men with low DFI whounderwent ICSI from 2007 to 2012. The data collectedincluded age of the male and female partners, semen analysisresults, DFI, woman's AMH level, the number of eggsretrieved, and the fertilization rate. The clinical pregnancyrates were calculated separately for fresh embryo transfersand frozen embryo transfers as well as cumulative rates.Clinical pregnancy was defined as the ultrasonographic
TABLE 1
Population analysis: men with a markedly high DNA fragmentation index
Outcome <15% DFI (n [ 114)
Male age (y) 38.4 � 0.45Female age (y) 35.18 � 4.13AMH (pmol/L) 20.26 � 24.73DFI (%) 10.4 � 2.7Sperm concentration (M/mL) 81.07 � 4.51Motility (/s) 53.32 � 1.36TME 135.3 � 115Note: AMH ¼ antim€ullerian hormone; TME ¼ total motile sperm per ejaculate.
Dar. ICSI for patients with very high DFI. Fertil Steril 2013.
76
visualization of one or more gestational sacs at the 6- to7-week ultrasound, and the miscarriage rate was determinedafter a clinical pregnancy had been established. Biochemicalpregnancies were not included in the analysis.
Sperm DNA damage was evaluated by flow cytometryand expressed as DFI using methods previously describedelsewhere (6). The patients were stratified by age and AMHlevels to exclude ovarian reserve confounding issues.
The data were analyzed using either GraphPad Prismversion 5.04 (GraphPad Software Inc.) or SAS version 9.2(SAS Institute). Descriptive statistics were performed usingthe mean and standard deviation to summarize continuousvariables, and counts and percentages to summarize categor-ical variables. Bivariate analyses were run to compare groupdifferences (high versus low DFI) with two sample two-sidedt tests looking at differences in continuous variables andchi-square or Fisher's exact tests looking at differences incategorical variables. Multivariable logistic regression modelswere run on the binary outcomes and looked at group differ-ences, controlling for confounders of interest. Outcomes suchas the miscarriage rate did not have a sufficient number ofevents to permit a multivariable model to be completed.
RESULTSPopulation Analysis
The mean DFI in the high DFI (group 1) was 60.3 � 9.2%(n ¼ 39) and the mean in the low DFI (group 2) was 10.4%� 2.7% (n ¼ 114). Men with high DFI were slightly olderthan the men with the low DFI, 40.9 � 0.92 years and38.4 � 0.45 years, respectively (P¼ .008). The mean spermconcentration in group 1 was 20.86� 2.82M/ml and in group2 was 81.07 � 4.51 M/mL (P< .0001), and the meanmotility in group 1 was 24.32 � 1.68 % and in group 2 was53.32 � 1.36 % (P< .0001). Statistically significant differ-ences were also observed in the number of total motile spermper ejaculate (TME) (Table 1). There was no statistically signif-icant difference in the average female age between the lowand high DFI groups overall or at any of the AMH levelswith the exception of the medium AMH (10–15) group.At the medium AMH level, the women in the high DFIgroup were older, 38.6 � 0.57 years versus 36.3 � 0.49years (P¼ .015). Smoking was similarly prevalent amonggroups 1 and 2 (17% and 16%, respectively) and among the
stratified AMH groups (low 12.5%, medium 16.6%, high11.1%, and very high 12.5%). There was no difference insmoking habits in comparing the pregnant with thenonpregnant couples in groups 1 and 2 (29% vs. 10%,P¼ .378; and 23% vs. 35%, P¼ .388, respectively), and nodifference in smokers among the couples who miscarried ingroup 1 versus 2 (50% vs. 38% P¼1).
Fertilization Rate/Oocyte and Embryo Results
The fertilization rate did not statistically significantly differbetween the DFI groups irrespective of AMH level (Fig. 1).
Pregnancy and Miscarriage Rates
The pregnancy rates after fresh embryo transfer and thecumulative pregnancy rates in the high versus low DFI groupswere not significantly different, and the same was true for theAMH stratification levels (Table 2). The miscarriage rate wasnot statistically statistically different between the groups,although a trend could be seen between the high versus thelow DFI groups. The rate was higher in the high DFI groupbut not significantly so across all AMH stratification levelswith the exception of the high AMH level, where a trendtoward a lower miscarriage rate was observed in the highDFI group over the low DFI group. However, none of thesedifferences reached statistical significance (Table 3).
We compared the DFI levels between patients with anongoing pregnancy to patients that experienced amiscarriagewithin groups 1 and 2. The DFI was not significantly differentin this analysis: group 1, 61.5 � 9.6 versus 65.8 � 12.7,respectively (P¼ .150); group 2, 10.2 � 2.3 versus 10.7 �2.9 (P¼ .522).
Bivariate and Multivariate Analysis
Bivariate analyses were run to compare differences in overallpregnancy and miscarriage rates in the high versus low DFI
FIGURE 1
Fertilization rates at each antim€ullerian hormone (AMH) level andoverall. No statistically significant differences were found in thefertilization rate when comparing the low and high DNAfragmentation index groups (P>.05).Dar. ICSI for patients with very high DFI. Fertil Steril 2013.
VOL. 100 NO. 1 / JULY 2013
groups with respect to the following parameters: femaleage, male age, AMH level, and total number of matureoocytes. There was no statistically significant differencefound between the two groups in terms of pregnancy rate ormiscarriage rate with respect to any of the other variables.Women who got pregnant tended to be younger, averageage 34.54 � 4.42 years, whereas the average age of thewomen who did not conceive was 36.25 � 3.93 years(P¼ .014). We also found that there was a higher average ofmature oocytes, 11 � 7.03, in cycles that resulted inpregnancy than in cycles that did not result in a pregnancy,6.10 � 4.43 (P< .001).
Multivariable linear or logistic regression models wererun on fertilization, fresh pregnancy, and total pregnancyrates between the groups, controlling for the confounders:female age, male age, AMH level, and number of matureoocytes. There was no statistically significant differencefound in the fertilization rate (73% vs. 67%, P¼ .35), the freshpregnancy rate (52% vs. 41%, (P¼ .42), or the total pregnancyrate (46% vs. 39%, P¼ .57) when controlling for theconfounding variables.
DISCUSSIONAssessing the male contribution to infertility for a couple isnot an easy task. Although the standard sperm analysisprovides important information, in many fertility clinicsadditional tests are performed, including DNA integritytesting. In our study, we compared patients with a markedlyhigh DFI with those with a low DFI, all undergoingIVF-ICSI.
Numerous cutoffs associated with flow-cytometry-basedDFI have been reported in the literature. The low level of DNAdamage <15% is well established. This level has been shownto be associated with excellent fertility potential. Low fertilitypotential is associated with DFI >30% (6), but this thresholdremains controversial. Currently no established absoluteupper limit that prevents a successful pregnancy exists.Reports on the effects of very high DFI levels (>50%) arelacking in the literature. Our hypothesis was that by includingsamples with at least 50% damaged spermatozoal DNAwe would be able to better determine the effect of DFI onpregnancy outcome.
In our study, we did not find statistically significant dif-ferences in fertilization or pregnancy rates with a very highDFI, which is in agreement with some previously publishedpapers using lower cutoffs. Esbert et al. (16) published find-ings of a prospective study of 178 couples. In their study,62 IVF and 116 ICSI cycles were analyzed both with own (n¼ 77) and donor (n ¼ 101) oocytes. The DNA fragmentationwas evaluated by terminal deoxynucleotidyl transferase-mediated dUDP nick-end labeling assay (TUNEL). They foundthat DNA fragmentation was not correlated with fertilizationrates in either IVF (r ¼ 0.08) or ICSI (r ¼ 0.04) cycles. In ad-dition, they showed that DNA fragmentation was similar incases where a pregnancy resulted and in cases where thewomen did not conceive. They also found that DNA fragmen-tation was similar in cases where miscarriages occurred and incases with successful deliveries. They also assessed whether
77
TABLE 2
Pregnancy rates: men with a markedly high DNA fragmentation index (DFI) compared with those with a low DFI.
Dar. ICSI for patients with very high DFI. Fertil Steril 2013.
ORIGINAL ARTICLE: ANDROLOGY
using a 36% sperm DNA fragmentation threshold wouldchange the outcome, and they found that it did not (16).
Borini et al. (17) retrospectively reviewed 132 patientswho underwent an IVF or ICSI cycles. The patients weredivided into low (<10) and high (>10) TUNEL. In the IVFgroup, no differences were found in clinical pregnancy ormiscarriage rates, whereas in the ICSI group the low TUNELgroup had a higher pregnancy rate and a lower miscarriagerate. They went on to set a 20% TUNEL threshold as a cutoff,but this did not change the results. Sample size and fragmen-tation threshold as well as combining IVF and ICSI were thelimitations of their study. To address those limitations, weset a very high DNA damage threshold, above 50% (mean60.3% � 9.2%), to help us detect whether a certain level ofdamage prevents fertilization with ICSI.
Antim€ullerian hormone is a good predictor for ovarianreserve, and its role in our study was primarily to help controlfor female factor infertility. However, AMH has limited utilityfor the prediction of pregnancy because AMH levels do notpredict egg quality, a factor that is highly influenced by thewomen's age (18, 19). Young women with low AMH levelswill have lower ovarian reserve compared with their agegroup, but they will have higher pregnancy rates comparedwith older women. In our study, the pregnancy rates did notappear to correlate with AMH levels, but the sample size ineach subgroup was too small to make a meaningfulcomparison. Also, the cohort consisted of relatively young
TABLE 3
Miscarriage rates: men with a markedly high DNA fragmentationindex (DFI) compared with those with a low DFI.
AMH levelLow DFI(<15%)
High DFI(>50%)
Pvalue
Low (%10) 20.0% (3/15) 66.6% (2/3) .17Medium (10<15) 25.0% (2/8) 50.0% (1/2) .49High (15–30) 37.5% (6/16) 30.0% (3/10) .70Very high (>30) 14.3% (2/14) 25.0% (1/4) .61All 24.5% (13/53) 36.8% (7/19) .37Dar. ICSI for patients with very high DFI. Fertil Steril 2013.
78
women. As stated earlier, we found no statisticallysignificant differences in fertilization, pregnancy, ormiscarriage rates.
Avenda~no (20) prospectively studied 36 ICSI cycles, inwhich low DNA damage was found to be a statisticallysignificant predictor of pregnancy. Our data contradicts theirresults, as we found no difference in the pregnancy rate afterICSI.
A recent meta-analysis, investigating the effect of spermDNA fragmentation on miscarriage rates, included 16 cohortstudies (2,969 couples), 14 of which were prospective. Thatanalysis showed a statistically significant increase in miscar-riage rate for patients with high DNA damage compared withthose with low DNA damage: relative risk (RR) ¼ 2.16 (1.54,3.03, P< .00001). Because different assays and differentthresholds for DNA damage were used, they also performeda subgroup analysis, which interestingly showed that themiscarriage association was strongest for the TUNEL assay:RR ¼ 3.94 (2.45, 6.32, P< .00001) (21). To evaluate spermDNA damage or fragmentation in the form of single ordouble-strand DNA breaks, several methods have beendeveloped of which the most commonly used is the spermchromatin structure assay (SCSA), expressed as DFI andTUNEL. The DFI reflects the susceptibility of sperm DNA topartial acid-induced denaturation, as determined by themetachromatic properties of acridine orange (6). The TUNELassay is based on the detection of dUTP at both single- anddouble-stranded DNA breaks by an enzymatic reaction (22).Although these tests are methodologically different, theyboth provide a direct evaluation of DNA damage. A statisti-cally significant correlation between DFI and TUNEL hasbeen reported (23).
Oocytes have the ability to repair damaged sperm.Depending on the damage, the repair could be fully successful(leading to a normal pregnancy), unsuccessful (leading tofailed fertilization/implantation), or partially successful.With partial repair, implantation may occur but may lead toa later miscarriage (19). This may partially explain the trendtoward higher miscarriages with elevated DNA damage. Inthe case of high DFI together with high AMH levels, the trend
VOL. 100 NO. 1 / JULY 2013
Fertility and Sterility®
toward a lower miscarriage rate that was detected could beattributed partially to better DNA repair mechanisms in thatsubgroup.
Miscarriages are multifactorial in origin and could resultfrom lower egg quality, as seen commonly in the infertilepopulation. Lekamge et al. (24) found a higher incidence ofmiscarriage during fresh embryo transfers when high andlow AMH concentrations were compared, ultimatelyculminating in a halving of the pregnancy rate per IVF cycle.Therefore, controlling for ovarian reserve is an importantparameter to include in an analysis with these types ofstudies. In our study, we compared the fertilization, clinicalpregnancy, and miscarriage rates after ICSI in men witha markedly high sperm DFI (>50%) versus those with a lowDFI, and we controlled for female factors by use of age andAMH level. We showed a trend toward an increase in thenumber of miscarriages that did not reach statisticalsignificance.
We realize that our study had a few limitations. First, thepopulation size of the patients with a very high DFI (>50%)was relatively small. A post hoc sample size calculation wasperformed to determine the number required to detect a statis-tically significant difference in the proportion of miscarriagesbetween the low and high DFI groups. The effect size used forthe calculation was a relative risk (RR) of 2.16, as found in themeta-analysis by Robinson et al. (21). The calculation foundthat we would have needed 42 pregnancies per group to detecta 2.16 RR at 80% power and alpha of 0.05 given an expected25% miscarriage rate in the low DFI group. Second, the highDFI group was significantly different from the low DFI, witha lower sperm count and lower motility in the high DFI group.These differences between the two groups might reflect animportant, untreated male factor in the high DFI group, whichcould impact IVF-ICSI success. Despite these differencesbetween the groups, the fertilization and pregnancy rateswere similar. The low DFI group included couples withunexplained infertility and couples who had failed multipleIUI cycles; thus, ICSI was chosen over standard IVF. As shownby Maman et al. (25), the ongoing pregnancy rate improveswhen ICSI is used for low responders. Previous low ovarianresponse, advanced maternal age, and previous poor fertiliza-tion with IVF were also indications for us to use ICSI in thesecouples.
When performing ICSI, embryologists attempt to selectspermatozoa with normal morphology and motility charac-teristics. It has been shown that preselected sperm tend tohave lower DNA damage (26, 27). This suggests thatperforming ICSI using carefully selected sperm is themethod of choice in patients with high DNA damage. Thetrend we observed toward an increase in miscarriages, asstatistically confirmed with the meta-analysis performed byRobinson et al., may suggest that any potential negativeimpact of a high DFI appears later in development (21).
Our study found no difference in the fertilization andimplantation rates or the pregnancy rates, but the trendobserved of an increase in the miscarriage rate deservesfurther study. Future studies should involve a larger cohortto determine whether this trend reaches statistical signifi-cance, and they should determine the exact mechanism of
VOL. 100 NO. 1 / JULY 2013
the effect of sperm DNA damage on the embryo as well astry to find methods to improve sperm selection for ICSI. Inaddition, long-term follow-up studies of the offspring frompregnancies conceived in the presence of a high sperm DFIwill be important for detecting any potential long-termdevelopmental problems.
Acknowledgments: The authors thank the staff at theCReATe Fertility Centre for their help with gathering all thecharts of the patient involved in the study, specifically DindoDacanay, B.Sc., and Jonathan Zicherman, B.Sc., for theirefforts with data collection, and also Alex Kiss, statisticianat Sunnybrook Health Sciences Centre, for help with thestatistical analysis of the data collected.
REFERENCES1. Hull MG, Glazener CM, Kelly NJ, Conway DI, Foster PA, Hinton RA, et al.
Population study of causes, treatment, and outcome of infertility. Br MedJ (Clin Res Ed) 1985;291:1693–7.
2. ChungCC, Fleming R, JamiesonME, Yates RW, Coutts JR. Randomized com-parison of ovulation induction with and without intrauterine insemination inthe treatment of unexplained infertility. Hum Reprod 1995;10:3139–41.
3. Lewis SE. Is sperm evaluation useful in predicting human fertility?Reproduction 2007;134:31–40.
4. Aitken RJ, De Iuliis GN. Value of DNA integrity assays for fertility evaluation.Soc Reprod Fertil Suppl 2007;65:81–92.
5. Cohen-Bacrie P, Belloc S, Menezo YJ, Clement P, Hamidi J, Benkhalifa M.Correlation between DNA damage and sperm parameters: a prospectivestudy of 1,633 patients. Fertil Steril 2009;91:1801–5.
6. Evenson DP, Kasperson K, Wixon RL. Analysis of sperm DNA fragmentationusing flow cytometry and other techniques. Soc Reprod Fertil Suppl 2007;65:93–113.
7. Zini A, Boman JM, Belzile E, Ciampi A. Sperm DNA damage is associatedwith an increased risk of pregnancy loss after IVF and ICSI: systematic reviewand meta-analysis. Hum Reprod 2008;23:2663–8.
8. Morris ID, Ilott S, Dixon L, Brison DR. The spectrum of DNA damage in humanspermassessedby single cell gel electrophoresis (Cometassay) and its relation-ship to fertilization and embryo development. Hum Reprod 2002;17:990–8.
9. Spano M, Bonde JP, Hjollund HI, Kolstad HA, Cordelli E, Leter G. Spermchromatin damage impairs human fertility. The Danish First PregnancyPlanner Study Team. Fertil Steril 2000;73:43–50.
10. Carrell DT, Liu L, Peterson CM, Jones KP, Hatasaka HH, Erickson L, et al.Sperm DNA fragmentation is increased in couples with unexplainedrecurrent pregnancy loss. Arch Androl 2003;49:49–55.
11. Cooke MS, Evans MD, Dizdaroglu M, Lunec J. Oxidative DNA damage:mechanisms, mutation, and disease. FASEB J 2003;17:1195–214.
12. Fraga CG, Motchnik PA, Shigenaga MK, Helbock HJ, Jacob RA, Ames BN.Ascorbic acid protects against endogenous oxidative DNA damage inhuman sperm. Proc Natl Acad Sci USA 1991;88:11003–6.
13. Basatemur E, Sutcliffe A. Follow-up of children born after ART. Placenta2008;29(Suppl B):135–40.
14. Ji BT, Shu XO, Linet MS, Zheng W, Wacholder S, Gao YT, et al. Paternalcigarette smoking and the risk of childhood cancer among offspring ofnonsmoking mothers. J Natl Cancer Inst 1997;89:238–44.
15. Sorahan T, Prior P, Lancashire RJ, Faux SP, Hulten MA, Peck IM, et al.Childhood cancer and parental use of tobacco: deaths from 1971 to1976. Br J Cancer 1997;76:1525–31.
16. Esbert M, Pacheco A, Vidal F, Florensa M, Riqueros M, Ballesteros A, et al.Impact of sperm DNA fragmentation on the outcome of IVF with own ordonated oocytes. Reprod Biomed Online 2011;23:704–10.
17. Borini A, Tarozzi N, Bizzaro D, Bonu MA, Fava L, Flamigni C, et al. SpermDNA fragmentation: paternal effect on early post-implantation embryodevelopment in ART. Hum Reprod 2006;21:2876–81.
18. Broer SL, van Disseldorp J, Broeze KA, Dolleman M, Opmeer BC, Bossuyt P,et al. Added value of ovarian reserve testing on patient characteristics in the
79
ORIGINAL ARTICLE: ANDROLOGY
prediction of ovarian response and ongoing pregnancy: an individual patientdata approach. Hum Reprod Update 2013;19:26–36.
19. Broer SL, Mol B, Dolleman M, Fauser BC, Broekmans FJ. The role ofanti-Mullerian hormone assessment in assisted reproductive technologyoutcome. Curr Opin Obstet Gynecol 2010;22:193–201.
20. Avenda~no C, Franchi A, Duran H, Oehninger S. DNA fragmentation ofnormal spermatozoa negatively impacts embryo quality and intracytoplas-mic sperm injection outcome. Fertil Steril 2010;94:549–57.
21. Robinson L, Gallos ID, Conner SJ, Rajkhowa M, Miller D, Lewis S, et al.The effect of sperm DNA fragmentation on miscarriage rates: a systematicreview and meta-analysis. Hum Reprod 2012;27:2908–17.
22. Gorczyca W, Traganos F, Jesionowska H, Darzynkiewicz Z. Presence of DNAstrand breaks and increased sensitivity of DNA in situ to denaturation inabnormal human sperm cells: analogy to apoptosis of somatic cells.Exp Cell Res 1993;207:202–5.
80
23. Barroso G, Morshedi M, Oehninger S. Analysis of DNA fragmentation,plasma membrane translocation of phosphatidylserine and oxidative stressin human spermatozoa. Hum Reprod 2000;15:1338–44.
24. Lekamge DN, Barry M, Kolo M, Lane M, Gilchrist RB, Tremellen KP.Antim€ullerian hormone as a predictor of IVF outcome. Reprod BiomedOnline 2007;14:602–10.
25. Maman E, Geva LL, Yerushalmi G, Baum M, Dor J, Hourvitz A. ICSIincreases ongoing pregnancy rates in patients with poor response cycle: mul-tivariate analysis of 2819 cycles. Reprod Biomed Online 2012;25:635–41.
26. Gandini L, Lombardo F, Paoli D, Caruso F, Eleuteri P, Leter G, et al. Full-termpregnancies achieved with ICSI despite high levels of sperm chromatindamage. Hum Reprod 2004;19:1409–17.
27. Hazout A, Dumont-Hassan M, Junca AM, Cohen Bacrie P, Tesarik J. High-magnification ICSI overcomes paternal effect resistant to conventionalICSI. Reprod Biomed Online 2006;12:19–25.
