RBMOnline - Vol 12. No Reproductive BioMedicine Online; on web 28 February 2006

Transcription

1 RBMOnline - Vol 12. No Reproductive BioMedicine Online; on web 28 February 2006 Meta-analyses were conducted to investigate the relationship of sperm DNA fragmentation on pregnancy outcome using in-vivo fertilization, IUI, routine IVF and ICSI. Couples with no known infertility problems were 7.0 times (CI 3.17, 17.7) more likely to achieve a pregnancy/delivery if the DNA fragmentation index (DFI) was <30% (n = 362, P = ) using in-vivo fertilization. Infertile couples using IUI were 7.3 times (CI 2.88, 18.3) more likely to achieve a pregnancy/delivery if their DFI was <30% (n = 518, P = ). With routine IVF, infertile couples were ~2.0 times (CI 1.02, 2.84) more likely to become pregnant if their DFI was <30% (n = 381, P = 0.03). For ICSI and/or routine IVF, the results showed a non-significant trend where infertile couples were 1.6 times (CI 0.92, 2.94) more likely to achieve a pregnancy/delivery if the DFI was <30% (n = 323, P = 0.06). The in-vivo and IUI meta-analyses were similar, indicating that IUI infertility patients with <30% DFI have as good a statistical probability of obtaining a pregnancy/delivery as in-vivo presumably fertile couples with the same DFI. These meta-analyses show that the Sperm Chromatin Structure Assay infertility test was significantly predictive for reduced pregnancy success using in-vivo, IUI and routine IVF, and to a lesser extent ICSI fertilization. Keywords: male infertility, meta-analysis, SCSA, sperm DNA fragmentation The DFI (DNA fragmentation index, formerly known as COMP α t ) threshold for infertility was first established in the comprehensive Georgetown Male Factor Infertility Study (Evenson et al., 1999; Zinaman et al., 2000). This study included 144 couples trying to conceive naturally. Sperm Chromatin Structure Assay (SCSA) data from male partners of 73 couples (group 1) achieving pregnancy during months 1 3 were used as the standard of sperm DNA integrity compatible with high fertility. Group 1 had significantly different DFI values compared with 40 couples (group 3) achieving pregnancy in months 4 12 (P < 0.01) and 31 couples (group 4) not achieving pregnancy (P < 0.001). In-vivo couples in this study were 6.5 times more likely to achieve a pregnancy when their DFI was below 30% (see Table 1 for a list of all odds ratios). In general, the SCSA data for individuals were highly repeatable from month to month. However, one couple s SCSA values were 64.4, 34.4 and 8.0% for months 1 3 respectively; no pregnancy occurred in months 1 and 2, but conception occurred in month 3. Data from this study were used to establish the statistical thresholds for the SCSA analysis of 30% DFI for significant lack of, >15 to <30% DFI for reasonable and 15% DFI for high fertility potential status. It is very important to note that this study clearly showed that some men with a >30% DFI fathered pregnancies; however, the time to pregnancy was longer for those conceiving in months 4 12 with higher mean DFI values. Spano et al. (2000) found in-vivo couples were 10 times more likely to achieve a pregnancy when DFI 40%. When IUI was the method of fertilization, Bungum et al. (2004) found couples were 8.7 times more likely to deliver a baby when DFI was 27% and high DNA stainability (HDS), the SCSA measurement associated with immature chromatin, was 10%. These results were in agreement with a later study presented at ESHRE 2005 by the same group. For routine IVF, Henkel et al. (2003, 2004) found a significant decrease in pregnancy

2 when sperm DNA fragmentation was >36.5% [TdT (terminal deoxynucleotidyl transferase)-mediated dudp nick-end labelling (TUNEL)]. Intracytoplasmic sperm injection (ICSI) appears to significantly overcome some negative effects of elevated sperm DNA fragmentation (Bungum et al., 2004; Virro et al., 2004); however, elevated sperm DNA fragmentation is of concern due to the association between assisted reproductive techniques and increased rates of birth defects (Fertility and Sterility Controversy, November 2005). The relationship of elevated sperm DNA fragmentation and increased rate of birth defects is shown by the trend for increased spontaneous abortions in several ICSI studies where pregnancies were obtained even though the sperm DNA fragmentation was elevated (Evenson et al., 1999; Carrell et al., 2003; Virro et al., 2004; Check et al., 2005; Zini et al., 2005). This relationship is also shown by the significant 8.7 OR of SCSA values and negative pregnancy outcome using IUI fertilization (Bungum et al., 2004). In the Controversy section of Fertility and Sterility (November, 2005), Olson et al. found IVF infants had an increased rate of major birth defects (in agreement with Hansen et al., 2002; Klemetti et al., 2005; Schieve et al., 2005), whereas no significant increases in major malformations were found when couples were treated with IUI. The objective of this paper was to evaluate by metaanalysis SCSA data and pregnancy outcome via four different fertilization methods. The SCSA analysis measures 5000 individual spermatozoa per sample using the statistically rigorous power of flow cytometry instrumentation (Evenson et al., 1980, 1999). Spermatozoa are treated with a low ph buffer that denatures DNA at sites of DNA strand breaks. The sample is then stained with acridine orange, a metachromatic dye that intercalates into native, double-stranded DNA and fluoresces green, while associating with single stranded DNA that fluoresces red. Based upon the percentage of spermatozoa with fragmented DNA, men are placed into statistical categories of: 15% DFI = excellent sperm DNA integrity, >15 to <30% DFI = good sperm DNA integrity and 30% DFI = fair to poor sperm DNA integrity. Note the change of terminology where previously (Evenson et al., 1999) the classifications were excellent, good and fair-topoor fertility potential. This was incorrect because the SCSA measures only infertility. Thus, current reports are grouped as excellent, good, and fair-to-poor sperm DNA integrity. The studies for the meta-analyses were selected from published studies and personal communications that used the SCSA analysis for the determination of sperm DNA fragmentation. Studies from Morris et al. (2002) (Comet analysis) and Henkel et al. (2004) (TUNEL analysis) were included because of the close correlations between the SCSA versus the Comet and TUNEL assays. The selection criteria for the in-vivo meta-analysis were presumably-fertile patients wishing to achieve pregnancy and who were discontinuing contraception. The selection criteria for the IUI, routine IVF and ICSI meta-analyses were couples undergoing infertility treatment and using SCSA, TUNEL or Comet for the determination of sperm DNA fragmentation. None of the selected studies listed female factors as a cause of possible infertility. Software used was SAS copyright 1999 by SAS Institute Inc., Cary, NC, USA; proprietary Software Version 8. The Cochran Mantel Haenszel statistic was used to calculate the metaanalyses, PROC FREQ. The Breslow Day test was used to test the homogeneity of the odds ratios. The Yates correction factor was used for the Larson et al., (2000) and the Larson-Cook et al., (2003) studies instead of the Chi-squared analysis because of the zero value for pregnancies above 30%. A meta-analysis was conducted to investigate the relationship of sperm DNA fragmentation to pregnancy outcome using IVF (Evenson et al., 1999; Spano et al., 2000). Using the Cochran Mantel Haenszel (CMH) statistic, the meta-analysis indicated that patients were 7.0 times (CI 3.17, 17.70) more likely to achieve a pregnancy/delivery if the DFI was <30% (n = 362, P = ). The Breslow Day test (BDT) was done to test the homogeneity of all the odds ratios for IVF used in the meta-analysis. The odds ratio for all studies tested was not significantly different (Chi-squared = 0.06, P=0.81) and showed similar trends (Table 1). When IUI was the method of fertilization, patients were 7.3 times (CI 2.88, 18.3) more likely to achieve a pregnancy/ delivery if the DFI < 30% (n = 518, P = ) (Bungum et al., 2004, 2005). Using the BDT, the odds ratio for all studies tested was not significantly different (Chi-squared = 0.04, P=0.84) and showed similar trends. When routine IVF was considered, couples were ~2.0 times (CI 1.02, 2.84) more likely to become pregnant if their DFI was <30% (n = 381, P = 0.03) (Larson-Cook et al., 2003; Adams et al., 2004; Bungum et al., 2004; Henkel et al., 2004). The odds ratio for all studies tested was not significantly different (Chi-squared = 5.26, P = 0.15) and showed similar trends. A meta-analysis of six studies using ICSI/and or routine IVF was also done to investigate the relationship of sperm DNA fragmentation on pregnancy outcomes (Larson et al., 2000; Morris et al., 2002; Larson-Cook et al., 2003; Bungum et al., 2004; Chohan, 2004; Check et al., 2005). The meta-analysis (CMH) indicated a trend where patients were 1.6 times (CI 0.92, 2.94) more likely to achieve a pregnancy/delivery if the DFI was <30% (n = 323, P=0.06). The BDT showed that the odds ratio for all studies tested were not significantly different (Chi-squared = 7.97, P = 0.16) and showed similar trends. The above meta-analyse: show that the SCSA infertility test is significantly predictive for reduced pregnancy success using in-vivo, IUI, and routine IVF and to a lesser extent ICSI fertilization. Morris et al. (2002) described their data as populations of high and low DNA damage. Eight samples fell between the high and low distinctions and were not included in the meta-analysis.

3 Table 1. Summary of the studies used for the meta-analysis investigating the relationship between sperm DNA fragmentation and pregnancy outcome. Author Odds ratio % DFI Assisted No. of χ 2 a Analysis reproduction patients procedure Evenson et al. (1999) 6.5 <30 In-vivo SCSA Spano et al. (2000) 10 <40 In-vivo 215 N/A SCSA Bungum et al. (2004) IUI SCSA Bungum (2005) IUI SCSA Larson-Cook et al. (2003) 0.5 e <27 IVF 55 NS SCSA Henkel et al. (2004) ~ IVF TUNEL Henkel et al. (2003) ~ IVF TUNEL Virro et al. (2004) b 2 <30 IVF SCSA Bungum et al. (2004) 1.5 <27 IVF 109 NS SCSA Morris et al. (2002) 1.5 Low DNA IVF or ICSI 52 NS Comet damage Chohan et al. (2004) 2 <30 IVF or ICSI 52 NS SCSA, TUNEL, SCD f Bungum et al. (2004) 3.1 >27 ICSI versus IVF 66 NS SCSA Larson-Cook et al. (2003) 4.8 e <27 ICSI 26 NS SCSA Larson et al. (2000) 23.8 e <27 ICSI SCSA Check et al. (2005) 2.1 <30 ICSI 106 NS SCSA Elliot et al. (2004) 1.5 <30 ICSI 68 NS SCSA Adams et al. (2004) d 9.5 <30 IVF 50 c SCSA For ease of reading the table, an example for the first row is provided: e.g. a patient is 6.5 times more likely to become pregnant and/or deliver if <30% of the spermatozoa have fragmented DNA using IVF. DFI = DNA fragmentation index; ICSI = intracytoplasmic sperm injection; IUI = intrauterine insemination; SCSA = sperm chromatin structure assay; TUNEL = TdT (terminal deoxynucleotidyl transferase)-mediated dudp nick-end labelling; N/A = not available; NS = not statistically significant. a Results were significant at P < b Patients were two times more likely to carry a pregnancy >12 weeks if DFI was <30%. c Donor egg cycles. d Patients were 9.5 times more likely to carry a pregnancy with high-quality donor eggs when sperm DFI was <30%. e Odds ratios were calculated using Yates correction factor (see text). f Sperm chromatin dispersion test. As mentioned above, the Georgetown Male Factor Infertility Study established the statistical ranges of excellent ( 15% DFI), good (>15 to <30% DFI), and fair to poor ( 30% DFI) sperm DNA integrity. This study included 144 couples trying to conceive naturally. Odds ratios from this study indicated that couples were 6.5 times more likely to become pregnant when their DFI was <30%. Spano et al. (2000) found that in-vivo patients (n = 215) (with no previous knowledge of their fertility status) showed a declining pregnancy outcome when their DFI was >20% and negligible pregnancies when the DFI approached 40%. Patients were 10 times more likely to achieve a pregnancy when DFI was 40%. In a recent study by Bungum et al. (2004), the relationship between the results of SCSA data and outcomes of IUI, IVF and ICSI fertilization were investigated. Two SCSA parameters, DFI and high HDS, gave a high predictive value regarding the outcome of IUI. HDS spermatozoa are known (Tritle and Evenson, unpublished data) to have immature chromatin and other characteristics of immature spermatozoa. IUI patients were grouped by 27% DFI and HDS 10% (group 1) and >27% DFI or HDS > 10% (group 2). Using odds ratios, group 1 with DFI 27% were 8.7 times more likely to deliver a baby than group 2 with DFI > 27%. In a later study by Bungum et al. (2005), similar results were found. Patients undergoing IUI were 7.0 times more likely to become pregnant when their DFI was <27% (n = 387, P = ). Duran et al. (2002) found no pregnancies were initiated when sperm DNA fragmentation was >12% (TUNEL) in 119 IUI patients. SCSA data have been shown to be closely related to TUNEL data (r = 0.859, P < 0.001) (Gorzcyca et al., 1993), strongly suggesting that the SCSA technique adds acridine orange fluorochrome markers at the site of DNA strand breaks which are also the sites of the enzymatic addition of a fluorochrome with the TUNEL assay. Studies on animal spermatozoa have shown similar correlations between SCSA and TUNEL data for bulls (r = 0.78, P < 0.001) stallions (r = 0.65, P < 0.001) and rams (r = 0.84, P < 0.001) (Sailer et al., 1995a). At present, however, the SCSA is a more reliable and repeatable technique for precise repeatable determination of DNA integrity. This is in

4 agreement with Erenpreiss et al. (2004), who showed significant correlations between SCSA and TUNEL data (r = 0.63, P = 0.005). SCSA and TUNEL data were positively related (P < 0.05) when homozygous β-thalassaemia patients with iron overload were evaluated for sperm DNA damage (Perera et al., 2002). Although TUNEL and SCSA tests likely measure the same sites of sperm DNA damage, the threshold for TUNEL from various laboratories appear to differ significantly. Duran et al. (2002) found no pregnancies when the TUNEL threshold was >12%, while Benchaib et al. (2003) found a significant decrease in pregnancy rate at a >20% threshold. While thresholds may differ between laboratories, there may be consensus within the same laboratory as shown by Henkel et al. (2003, 2004), where a significant decrease in pregnancies was found when the TUNEL threshold was >36.5%. Henkel et al. (2004) demonstrated that DNA fragmentation, as determined by the TUNEL assay, is significantly predictive for pregnancy rate in IVF. Patients with a lower percentage of DNA fragmentation (TUNEL) had an almost 2 pregnancy rate (34.65 versus 19.05%, P = ), compared with patients with higher levels of DNA fragmentation. In this study, the receiver operating characteristic (ROC) established TUNEL threshold was set at 36.5% for TUNEL negative spermatozoa. Note that ROC analysis is not an appropriate statistical tool to analyse SCSA data. ROC analysis makes use of sensitivity and specificity, which is used for disease testing. The disease test gives a positive or negative (yes, the disease is present or no, it is not), which are absolute outcomes; whereas for the SCSA test, chances for pregnancy of >30% are reduced but there is a chance of pregnancy; likewise if the DFI is <30% there is no guarantee that a pregnancy will occur due to female and other issues. The purpose of ROC analysis is to establish a threshold for the absence or presence of a disease (in accordance with Henkel et al., 2003, 2004). Odds ratio is a more appropriate tool to analyse SCSA data. This is in agreement with an earlier study by Henkel et al. (2003), where IVF patients were ~2 times more likely to become pregnant when using ROC established TUNEL threshold set at 36.5% for TUNEL negative spermatozoa. When SCSA defined levels of sperm DNA fragmentation were <30%, routine IVF patients were twice as likely to carry a pregnancy >12 weeks (Virro et al., 2004). Bungum et al. (2004) found IVF patients (n = 109) in the SCSA defined <27% DFI group were 1.5 times more likely to deliver as the >27% DFI group. In these studies, high levels of DNA fragmentation were clearly shown to be predictive for reduced pregnancy in routine IVF. In ICSI patients (n = 66) the SCSA defined <27% DFI group were 2.5 times more likely to deliver a baby as the >27% DFI group (Bungum et al., 2004). Comparison of ICSI and routine IVF in the >27% group indicated that the ICSI patients were 3.6 times more likely to deliver than the routine IVF group. These results were significant using the chi-square test. Benchaib et al. (2003) found DNA fragmentation was significantly lower when a pregnancy was obtained using ICSI (P < 0.05, n = 54). The relationship between sperm DNA damage and pregnancy rate was investigated in 52 couples undergoing routine IVF and ICSI procedures (Chohan, 2004). The authors concluded that while there was no significant relationship with IVF outcome, the negative trend for increasing DFI and pregnancy outcome may become significant with a larger study population. Odds ratio calculations showed that a patient was twice times more likely to become pregnant when the DFI was <30% (not statistically significant). Using Comet analysis, Morris et al. (2002) found that elevated sperm DNA damage was predictive of embryo development failure in a group of IVF/ICSI men. Three of the nine pregnancies (33%) in the elevated sperm DNA damaged group resulted in an early miscarriage, while all six of the pregnancies in the low sperm DNA damaged group yielded live births. The aforementioned published studies and personal communications were chosen to be part of the meta-analysis because of the solidity of the data results in contrast to the results of Payne et al., (2005), which indicate that The poorer the integrity of sperm nuclear DNA, the better the pregnancy outcome. This is in contrast to the couple of hundred manuscripts on human and animal studies that, overall, conclude that the integrity of the sperm DNA as measured by SCSA, TUNEL, Comet-single cell gel electrophoresis and others is closely related to reproductive outcomes. Payne et al. wish to redefine the relationship between SCSA data and ART outcomes with 100 couples, which is an insufficient number to reach a valid conclusion that the SCSA (over 100,000 sperm samples measured in the laboratory) must be redefined. The 100 couples from the Payne et al. study included: 45 female infertility unspecified, 17 endometriosis, 17 tubal, 13 ovary polycystic ovary syndrome, 2 amenorrhoea, and 6 male infertility oligospermia. If these patients were the recipients of sperm populations in the <9% DFI group, this factor alone could explain why pregnancy rates were lower in the group with the best sperm DNA integrity. The SCSA predicts infertility, and not fertility. S Wood and C Adams (personal communication), have previously reported that DFI of <15% is associated with high assisted reproduction pregnancy rates (Adams et al., 2004), a finding that is consistent with many other studies in an array of settings (Evenson et al., 1999), but had never specifically examined outcomes from cycles in which DFI was below 10%. After a review of all assisted reproduction cycles over the past 3 years in which the precycle DFI was 9% or less, they found that in those 34 cycles the clinical pregnancy rate was 67.8%, much higher than the 9% reported by the Payne group. This high pregnancy rate is, of course, what would be expected based on previous studies in this area. The reasons for the unexpected results reported by Payne et al. are unclear, but are probably due to the small sample size and many confounding female factors. Recurrent pregnancy loss is a devastating issue for infertility patients. Current research indicates a trend towards increased spontaneous abortions when the DFI is 30%. In the Georgetown study, where human SCSA statistical thresholds were first derived, combined DFI and HDS scores predicted 39% of miscarriages (Evenson et al., 1999; Zinaman et al., 2000). Check et al. (2005) found slightly lower clinical pregnancies and higher miscarriage rates in the >30% DFI patient group resulting in an ongoing pregnancy rate that was almost half of the <15 and 15 30% groups though the differences were not significant. Zini et al. (2005), also found a trend for spontaneous abortions in couples with DFI >30%. A 30% DFI score was associated with a higher rate of spontaneous abortion at 12 weeks of gestation (P

5 < 0.01) in comparison to the <30% group (Virro et al., 2004). Men of couples with repetitive (>2) spontaneous abortions had a 4-fold increase in TUNEL positive spermatozoa relative to sperm donors (Carrell et al., 2003). Sperm DNA damage is multifactorial and may be due to many environmental conditions such as chemotherapy, radiation, some prescription medications, air pollution, smoking, pesticides, chemicals, heat, assisted reproduction preparation protocols (Evenson et al., 1980, 1985, 1986, 1989, 1991, 1993; Evenson and Jost, 1993; Sailer et al., 1995b; Fossa et al., 1997; Potts et al., 1999; Rignell-Hydbom et al., 2005; Rubes et al., 2005; Adams, personal communication) and various pathological conditions including cryptorchidism, cancer, fever, age, infection, leukocytospermia and varicocele among others (Evenson et al., 2000; Zini et al., 2001; Alvarez et al., 2002; Aitken et al., 2003; Burrello et al., 2004; Stahl et al., 2004). It would appear that elevated sperm fragmentation, damage due to reactive oxygen species and apoptotic events may all play a part in the pathogenesis of male infertility (Aitken et al., 2003). Protamine-deficient spermatozoa may also contribute to elevated sperm DNA fragmentation (Nasr-Esfahani et al. 2005). Several studies have shown a correlation between elevated protamine-1/protamine-2 (P1/P2) ratio and male infertility (de Yebra et al., 1993; Carrell and Liu, 2000). Aoki et al. (2005) found the correlation between elevated levels of sperm DNA fragmentation and low P1/P2 levels significant, but biologically weak. Decreased P1/P2 ratios were found in a population of infertile males undergoing IVF and/or ICSI. The overall fertilization rate (IVF and ICSI) was significantly decreased in patients with a low P1/P2 ratio compared with patients with normal and elevated P1/P2 ratios. Fertilization has shown weakly significant to non-significant correlations with elevated sperm DNA fragmentation (Henkel et al., 2004; Virro et al., 2004). The SCSA is a powerful statistical tool that can be used for statistical predictions on negative pregnancy outcomes with invivo, IUI and routine IVF fertilization. As stated in a current review by Erenpriess et al. (2006): at the moment SCSA is the only method that has demonstrated clear clinically useful cut-off levels between fertile and infertile men and its prognostic value for ART has also been shown The undisputed advantages of this technique are its robustness and small intraand inter-assay variations. For patients presenting with difficulties conceiving, multiple miscarriages, and high DFI values, the SCSA can be used to guide the patient towards ICSI fertilization or use the SCSA analysis as part of the initial infertility workup, as many physicians currently do. Reports have been conflicting regarding the correlations between SCSA DFI values and standard semen analysis measures. Results from several studies show r-values between DFI and classical semen measurements ranging from 0.42 to 0.12 for sperm density and 0.60 to 0.25 for morphology and for motility from 0.59 to While many of the DFI values and semen analysis measures are statistically significant, the correlations may have little biological meaning (Table 2). The great variability of the r values is probably due to variations in the measurement of the standard semen analysis. Furthermore, DFI showed no significant relationships to fertilization (Henkel et al., 2004, 2003; Virro et al. 2004), implantation rates (Bungum et al., 2004), and embryo quality (Benchaib et al., 2003). Since the male genome is not activated until day three, embryo quality at the time of transfer may not show future negative male genome effects. It is of the utmost importance to differentiate between the various assisted reproduction procedures to further define the role of the SCSA in infertility testing. Thus in-vivo, IUI, routine Table 2. Correlations between DNA fragmentation index values and sperm concentration, motility and morphology. All correlations are significant at P < 0.05 or lower except for those indicated as not statistically significant (NS). Concentration Motility Morphology Larson-Cook et al. (2003) Bungum et al. (2004): IUI NS 0.25 Bungum et al. (2004): IVF Significant Bungum et al. (2004): ICSI Significant Evenson et al. (1999): in vivo Spano et al. (2000): in vivo 0.23 NS 0.25 Virro et al. (2004): IVF/ICSI Sanchez-Pena et al. (2004) NS NS NS Rignell-Hydbom et al. (2005): exposure NS 0.37 NS to pentachlorobenzene and DDE Saleh et al. (2003): IUI, IVF, ICSI Zini et al. (2001) NS Giwercman et al. (2003): Danish 1st 0.53 Giwercman et al. (2003): Swedish military 0.38 Larson et al. (1999) ICSI = intracytoplasmic sperm injection; IUI = intrauterine insemination.

6 IVF and ICSI results must be reported separately to evaluate the effectiveness of the SCSA analysis. In summary, the data shown in this study clearly indicate that the SCSA is an important component of the infertility workup, in agreement with the current conclusions of Erenpreiss et al. (2006): Without doubt the existing data justify the necessity to introduce sperm DNA damage assessment into the routine infertility investigation. The similarity of the in-vivo and IUI meta-analyses indicate that for infertility patients undergoing IUI, their statistical chances of obtaining a pregnancy/delivery when DFI is <30% are just as encouraging as in-vivo presumably fertile couples with DFI <30. A DFI 30% does not preclude a normal full-term pregnancy. However, previous and current research continue to support the view that an approximately >30% DFI and >15% spermatozoa with the percentage of HDS spermatozoa (immature) in raw semen are statistically significant with regards to reduced pregnancy outcomes for in-vivo, IUI and routine IVF fertilizations as well as an increased risk for early spontaneous miscarriage. The data suggest that if a man has a DFI of >30% that IUI should probably not be considered and the couple move to routine IVF or preferably ICSI; however, patients should be counselled as to the risk of malformations associated with these assisted reproduction procedures. Research supported by in part by grant R from the US Environmental Protection Agency. This is South Dakota Agricultural Experiment Station publication #3517. Adams C, Anderson L, Wood S 2004 High, but not moderate, levels of sperm DNA fragmentation are predictive of poor outcome in egg donation cycles [abstract no. O-110]. In: Abstracts of the Scientific Oral and Poster Sessions, October 16 20, 2004, Philadelphia, USA. Fertility and Sterility 82 (suppl. 2), S44. Aitken RJ, Baker MA, Sawyer D 2003 Oxidative stress in the male germ line and its role in the aetiology of male infertility and genetic disease. Reproductive BioMedicine Online 7, Alvarez JG, Sharma RK, Ollero M et al Increased DNA damage in sperm from leukocytospermic semen samples as determined by the sperm chromatin structure assay. Fertility and Sterility 78, Aoki VW, Moskovtsev SI, Willis J et al DNA integrity is compromised in protamine-deficient human sperm. Journal of Andrology 26, Benchaib M, Braun V, Lornage J et al Sperm DNA fragmentation decreases the pregnancy rate in an assisted reproductive technique. Human Reproduction 18, Bungum M 2005 ESHRE abstracts, personal communication. Copenhagen, Denmark, June pp Bungum M, Humaidan P, Spano M et al The predictive value of sperm chromatin structure assay (SCSA ) parameters for the outcome of intrauterine insemination, IVF and ICSI. Human Reproduction 19, Burrello N, Calogero AE, Perdichizzi A 2004 Inhibition of oocyte fertilization by assisted reproductive techniques and increased sperm DNA fragmentation in the presence of Candida albicans: a case report. Reproductive BioMedicine Online 5, Carrell DT, Liu L 2001 Altered protamine 2 expression is uncommon in donors of known fertility, but common among men with poor fertilizing capacity, and may reflect other abnormalities of spermiogenesis. Journal of Andrology 22, Carrell DT, Liu L, Peterson CM et al Sperm DNA fragmentation is increased in couples with unexplained recurrent pregnancy loss. Archives of Andrology 49, Check JH, Graziano V, Cohen R, Krotec J, Check ML 2005 Effect of an abnormal sperm chromatin structural assay (SCSA ) on pregnancy outcome following (IVF) with ICSI in previous IVF failures. Archives of Andrology 51, Chohan KR, Fiffin JT, Lafromboise M et al Sperm DNA damage relationship with embryo quality and pregnancy outcome in IVF patients [abstract no. O-137]. In: Abstracts of the Scientific Oral and Poster Sessions, October 16 20, 2004, Philadelphia, USA. Fertility and Sterility 82 (suppl. 2), S55. de Yebra L, Ballesca JL, Vanrell JA et al Complete selective absence of protamine p2 in humans. Journal of Biological Chemistry 268, Duran EH, Morshedi M, Taylor S, Oehninger S 2002 Sperm DNA quality predicts intrauterine insemination outcome: a prospective cohort study. Human Reproduction 17, Elliot TA, Taylor TH, Jacobs CA et al A high DFI score does not preclude patients from achieving a pregnancy with ICSI [abstract no. O-117]. In: Abstracts of the Scientific Oral and Poster Sessions, October 16 20, 2004, Philadelphia, USA. Fertility and Sterility 82 (suppl. 2), S47. Erenpreiss J, Spano M, Erenpreisa J, Bungum M, Giwercman A 2006 Sperm chromatin structure and male fertility: biological and clinical aspects. Asian J Androl 8, Erenpreiss J, Jepson K, Giwercman A et al Toluidine blue cytometry test for sperm DNA conformation: comparison with the flow cytometric sperm chromatin structure and TUNEL assays. Human Reproduction 19, Evenson DP, Jost LK 1993 Hydroxyurea exposure alters mouse testicular kinetics and sperm chromatin structure. Cell Proliferation 26, Evenson DP, Jost LK, Corzett M, Balhorn R 2000 Characteristics of human sperm chromatin structure following an episode of influenza and high fever: a case study. Journal of Andrology 21, Evenson DP, Jost LK, Zinaman MJ et al Utility of the sperm chromatin structure assay (SCSA ) as a diagnostic and prognostic tool in the human fertility clinic. Human Reproduction 14, Evenson DP, Jost LK, Baer RK 1993 Effects of methyl methanesulfonate on mouse sperm chromatin structure and testicular cell kinetics. Environmental and Molecular Mutagenesis 21, Evenson DP, Jost LK, Baer RK et al Individuality of DNA denaturation patterns in human sperm as measured by the sperm chromatin structure assay. Reproductive Toxicology 5, Evenson DP, Baer RK, Jost LK 1989 Long-term effects of triethylenemelamine exposure on mouse testis cells and sperm chromatin structure assayed by flow cytometry. Environmental and Molecular Mutagenesis 14, Evenson DP, Baer RK, Jost LK 1986 Toxicity of thiotepa on mouse spermatogenesis as determined by dual-parameter flow cytometry. Toxicology and Applied Pharmacology 82, Evenson DP, Higgins PJ, Grueneberg D, Ballachey BE 1985 Flow cytometric analysis of mouse spermatogenic function following exposure to ethylnitrosourea. Cytometry 6, Evenson DP, Darzynkiewicz Z, Melamed MR 1980 Relation of mammalian sperm chromatin heterogeneity to fertility. Science 240, Fossa SD, De Angelis P, Kraggerud SM et al Prediction of posttreatment spermatogenesis in patients with testicular cancer by flow cytometric sperm chromatin structure assay. Cytometry 30,

7 Giwercman A, Richthoff J, Hjollund H et al Correlation between sperm motility and sperm chromatin structure assay parameters. Fertility and Sterility 80, Gorczyca W, Gong J, Darzynkiewicz Z 1993 Detection of DNA strand breaks in individual apoptotic cells by the in situ terminal deoxynucleotidyl transferase and nick translation assays. Cancer Research 53, Hansen M, Kurinczuk JJ, Bower C, Webb S 2002 The risk of major birth defects after intracytoplasmic sperm injection and in vitro fertilization. New England Journal of Medicine 346, Henkel R, Hajimohammad M, Stalf T et al Influence of deoxyribonucleic acid damage on fertilization and pregnancy. Fertility and Sterility 81, Henkel R, Kierspel E, Hajimohammad M et al DNA fragmentation of spermatozoa and assisted reproduction technology. Reproductive BioMedicine Online 7, Klemetti R, Gissler M, Sevon T et al Children born after assisted fertilization have an increased rate of major congenital anomalies. Fertility and Sterility 84, Larson KL, DeJonge CJ, Barnes AM et al Sperm chromatin structure assay parameters as predictors of failed pregnancy following assisted reproductive techniques. Human Reproduction 15, Larson KL, Brannian JD, Timm BK et al Density gradient centrifugation and glass wool filtration of semen remove spermatozoa with damaged chromatin structure. Human Reproduction 14, Larson-Cook K, Brannian JD, Hansen KA et al Relationship between the outcomes of assisted reproductive techniques and sperm DNA fragmentation as measured by the sperm chromatin structure assay. Fertility and Sterility 80, Morris ID, Ilott S, Dixon L, Brison DR 2002 The spectrum of DNA damage in human sperm assessed by single cell gel electrophoresis (Comet assay) and its relationship to fertilization and embryo development. Human Reproduction 17, Nasr-Esfahani MH, Salehi M, Razavi S et al Effect of sperm DNA damage and sperm protamine deficiency on fertilization and embryo development post-icsi. Reproductive BioMedicine Online 11, Olson CK, Keppler-Noreuil KM, Romitti PA et al In vitro fertilization is associated with an increase in major birth defects. Fertility and Sterility 84, Payne JF, Raburn DJ, Couchman GM et al Redefining the relationship between sperm deoxyribonucleic acid fragmentation as measured by the sperm chromatin structure assay and outcomes of assisted reproductive techniques. Fertility and Sterility 84, Perera D, Pizzey A, Campbell A et al Sperm DNA damage in potentially fertile homozygous beta-thalassaemia patients with iron overload. Human Reproduction 17, Potts RJ, Newbury CJ, Smith G et al Sperm chromatin damage associated with male smoking. Mutation Research 423, Rignell-Hydbom A, Rylander L, Giwercman A et al Exposure to PCBs and p,p -DDE and human sperm chromatin integrity. Environmental Health Perspectives 113, Rubes J, Selevan SG, Evenson DP et al Episodic air pollution is associated with increased DNA fragmentation in human sperm without other changes in semen quality. Human Reproduction 20, Saleh RA, Agarwal A, Nada EA et al Negative effects of increased sperm DNA damage in relation to seminal oxidative stress in men with idiopathic and male factor infertility. Fertility and Sterility 79 Suppl 3, Sailer BL, Jost LK, Evenson DP 1995a Mammalian sperm DNA susceptibility to in situ denaturation associated with the presence of DNA strand breaks as measured by the terminal deoxynucleotidyl transferase assay. Journal of Andrology 16, Sailer BL, Jost LK, Erickson KR et al. 1995b Effects of X-irradiation on mouse testicular cells and sperm chromatin structure. Environmental and Molecular Mutagenesis 25, Sanchez-Pena LC, Reyes BE, Lopez-Carrillo L et al Organophosphorous pesticide exposure alters sperm chromatin structure in Mexican agricultural workers. Toxicology and Applied Pharmacology 196, Schieve LA, Rasmussen SA, Reefhuis J 2005 Risk of birth defects among children conceived with assisted reproductive technology: providing an epidemiologic context to the data. Fertility and Sterility 84, Spano M, Bonde J, Hjollund HI et al Sperm chromatin damage impairs human fertility. Fertility and Sterility 73, Stahl O, Eberhard J, Jepson K et al The impact of testicular carcinoma and its treatment on sperm DNA integrity. Cancer 100, Virro MR, Larson-Cook KL, Evenson DP 2004 Sperm chromatin structure assay (SCSA ) related to blastocyst rate, pregnancy rate and spontaneous abortion in IVF and ICSI cycles. Fertility and Sterility 8, Zinaman MJ, Brown CC, Selevan SG, Clegg ED 2000 Semen quality and human fertility: a prospective study with healthy couples. Journal of Andrology 21, Zini A, Blumenfeld A, Libman J, Willis J 2005 Beneficial effect of microsurgical varicocelectomy on human sperm DNA integrity. Human Reproduction 20, Zini A, Bielecki R, Phang D, Zenzes MT 2001 Correlations between two markers of sperm DNA integrity, DNA denaturation and DNA fragmentation, in fertile and infertile men. Fertility and Sterility 75, Received 2 September 2005; refereed 13 September 2005; accepted 25 January 2006.