Keywords

1 Introduction

Sperm is a vehicle that aids the transport of the haploid paternal genome to the oocyte. The delivery of intact and complete genetic material to the oocyte is required for normal embryonic development [1]. To facilitate this process, the sperm nucleus is equipped with a unique design of nuclear architecture, where the nuclear proteins are replaced by smaller and positively charged protamines that allow the chromatin to form a compact structure [2]. During sperm nuclear structural reorganization , the sperm loses its cytoplasm, which provides the sperm its streamline nature and facilitates movement through the male and female reproductive tract. On the other hand, the lack of cytoplasm leaves the nucleus vulnerable and unprotected against the free radicals [3]. The seminal plasma not only acts as a medium for the sperm to swim in but also scavenges the free radicals to minimize the effect of oxidative stress-mediated DNA damage [4].

Despite such precaution, DNA fragmentation is common and is believed to be a property of all sperm. However, the level of DNA damage may vary from one sperm to another [5]. DNA damage occurring in sperm can be the result of intrinsic factors where poor organization of sperm chromatin leaves the sperm vulnerable to oxidative stress-mediated DNA damage [1]. Studies have suggested that there may be a cascade of events that starts with seminal oxidative stress leading to DNA base modifications and DNA fragmentation resulting in apoptosis of sperm [6, 7]. Other factors such as medication, heat, radiation, etc. are some of the extrinsic factors also known to cause sperm DNA damage [8,9,10,11]. Either way, sperm lacks any DNA repair mechanism to fix the fragmented DNA, and therefore the damage occurring to sperm DNA is believed to be an irreversible process.

Most commonly studied damage to the sperm chromatin are the single- and double-strand breaks, commonly known as DNA fragmentation. A number of tests are now available to measure the level of sperm DNA fragmentation rates. Of these methods, the single-cell gel electrophoresis (commonly called as the comet assay ) and the terminal deoxynucleotide transferase-mediated dUTP nick-end labeling (TUNEL) assays more directly measure the level of DNA fragmentation. Whereas, the sperm chromatin dispersion (SCD) assay (commonly called as the halo test ) and sperm chromatin structure assay (SCSA) are known to indirectly measure the level of DNA fragmentation in sperm. Each of these tests measure different aspects of DNA fragmentation in sperm.

2 Effect of Sperm DNA Fragmentation on Clinical Outcomes: Literature Review

The comprehensive review of the published literature yielded 88 articles for systematic review, following exclusion of overlapping data, inappropriate sampling method, assays that are less commonly used (neutral comet assay, in situ nick translation assay and acridine orange slide-based staining method), and studies with insufficient data. The studies included for systematic review involved DNA fragmentation assays using TUNEL assay (34 studies), SCSA (31 studies), comet assay (12 studies), and SCD assay (11 studies). Based on the treatment types, these studies involve IUI (11 studies), IVF (28 studies), ICSI (33 studies), and IVF+ICSI mixed (17 studies). 42 studies were identified that compared fertile and infertile men using the abovementioned assays. Sixty-seven of the abovementioned studies provided sufficient data to construct two-by-two table to perform a meta-analysis, and the data of meta-analysis was recently published [12]. Among these studies, there were differences in definition of threshold values for DNA damage assays, study design, lack of control for female factors, small sample size, diverse patient group, insufficient statistical power, non-consecutive recruitment of patients, and variations in the protocols used to measure DNA fragmentation, and in some studies the selection of subjects were not clearly stated.

2.1 Is Sperm DNA Fragmentation Associated with Male Fertility ?

A number of studies have compared DNA fragmentation between fertile and infertile men [13,14,15,16,17,18,19,20,21]. Overall, DNA fragmentation is more prevalent in the sperm of infertile men and may contribute to their declined fertility status. In addition, chromosomal abnormalities were shown to be increased in these patients [22, 23]. An increase in the level of DNA fragmentation in infertile men can be attributed to abnormal histone to protamine exchange [24], abnormal protamine content and ratio [25], and reduced antioxidant activity in the seminal plasma [26, 27]. A study examining the effect of DNA fragmentation on male fertility among first-pregnancy planners with no previous knowledge of their fertility capability suggested that fecundity declined with an increase in sperm DNA fragmentation, indicating the necessity of normal sperm chromatin for the expression of full male fertility potential [28]. Overall, the existing literature suggests a negative impact of sperm DNA fragmentation on male reproductive health.

2.2 Can Sperm DNA Fragmentation Predict Intrauterine Insemination (IUI) Success ?

We identified ten studies that analyzed the association between sperm DNA fragmentation with IUI outcome. A total of 1673 IUI cycles were analyzed using SCSA (seven studies), TUNEL (two studies), and SCD (one study) assays. The results from five of the seven studies by SCSA [18, 29,30,31,32] and one study using the TUNEL assay [33] suggested a significant statistical difference in the level of sperm DNA fragmentation between the clinically pregnant and nonpregnant groups. Conclusive results were not reported in two studies using SCSA [34, 35], while no correlations were reported in two studies: using TUNEL assay [36] and using SCD assay [37].

Data was available to construct a two-by-two table from six of the seven studies performed using SCSA (except for Alkhayal et al. [35]—data not available). A meta-analysis was performed on five studies (Bungum et al. [29]—overlapping study was excluded) consisting of 1135 IUI cycles and with an overall pregnancy rate of 18.23%, resulted in an odds ratio of 5.61 (CI: 2.59–12.16; Z statistics: 4.37; p < 0.0001) and relative risk of 1.17 (CI: 1.12–1.22; p < 0.0001) indicating a strong association between sperm DNA fragmentation and IUI outcome (unpublished data). The positive and negative predictive values were 96.0% and 15.9%, respectively. This model provided a low sensitivity (16%) but high specificity (93%) values. In conclusion, this meta-analysis showed a slight but significant predictive ability of DNA fragmentation assay (SCSA) to predict IUI success, and this is in contrast to recommendations provided by the Practice Committee of the American Society for Reproductive Medicine [38], which included four of the ten studies presented above for their analysis.

2.3 Is Sperm DNA Fragmentation Associated with Fertilization In Vitro ?

We identified 67 eligible articles that analyzed sperm DNA fragmentation with assisted reproductive treatment (ART) outcomes, of which 15 did not have data associating fertilization rate with sperm DNA fragmentation. The 52 eligible articles included 73 studies (25 IVF, 31 ICSI, and 17 mixed IVF+ICSI studies) and involved 8590 treatment cycles (2997 IVF, 2470 ICSI, and 3123 mixed IVF+ICSI cycles). Twenty-eight of the 73 studies (13 TUNEL, 5 SCSA, 6 SCD, and 4 comet) reported a significant inverse relationship between sperm DNA damage and fertilization rate, whereas the other 45 studies (16 TUNEL, 18 SCSA, 6 SCD, and 5 comet) showed no significant relationship between these parameters (Table 23.1).

Table 23.1 Summary of studies associating sperm DNA fragmentation with fertilization rate

Although we did not perform a formal meta-analysis, our systematic review of studies on sperm DNA fragmentation and fertilization rate after IVF and/or ICSI demonstrated that 38% (28/73) of the studies reported a significant inverse relationship between the two parameters. We found that a higher proportion of the IVF studies (60% or 16/25) reported a significant inverse relationship between sperm DNA fragmentation and fertilization rate than the ICSI (23% or 7/31) and mixed IVF+ICSI studies (35% or 6/17) (Table 23.1). A complete description of studies associating sperm DNA fragmentation with fertilization rate is presented in Table 23.2.

Table 23.2 Description of studies associating sperm DNA fragmentation with fertilization rate and embryo quality

It is known that sperm progressive motility and sperm DNA fragmentation are the two most important sperm factors to affect IVF rate [39]. Indeed, there is an association between sperm DNA fragmentation and progressive motility [40,41,42,43,44,45], and this may explain the influence of these two parameters on IVF rates. A possible explanation for fertilization failure could be that sperm with abnormal chromatin may not properly decondense after penetrating the oocyte, and this can prevent development of the pronuclear stage [46,47,48]. The differential adverse effect of sperm DNA fragmentation on IVF and ICSI fertilization may be due to the fact that conventional IVF occurs “naturally” as a result of sperm-oocyte interaction, whereas during ICSI treatment this natural selection process is bypassed [49]. With ICSI, the embryologist manually selects morphologically normal and motile sperm [50], which may increase the probability of selecting sperm with low DNA fragmentation [51], as negative correlations between these parameters have been reported [33, 40,41,42, 52,53,54,55,56,57]. The differential effect of sperm DNA fragmentation on IVF and ICSI fertilization may also be influenced by the atypical (delayed) sperm nuclear decondensation that occurs after ICSI [58]. These results presented here are also in agreement with the meta-analysis [59], showing that ICSI fertilization rates are higher than IVF rates in patients with unexplained infertility and in normozoospermic men presented with increased sperm DNA fragmentation [45, 60, 61]. In conclusion, sperm DNA fragmentation may be associated with IVF rate but not with ICSI fertilization rates.

2.4 Does Sperm DNA Fragmentation Decrease Embryo Quality ?

We identified 67 eligible articles that analyzed sperm DNA fragmentation with ART outcomes, of which 22 articles did not have data associating embryo quality with sperm DNA fragmentation. The 45 eligible articles included 62 studies (22 IVF, 24 ICSI, and 16 mixed IVF+ICSI studies) and involved 9055 treatment cycles (3957 IVF, 2409 ICSI, and 2689 mixed IVF+ICSI cycles). Embryo markers such as embryo grade (33 studies), embryo development (20 studies), fragmentation (3 studies), combined embryo grade and development (4 studies), multinucleation (1 study), and embryo cleavage rate (1 study) were used to determine embryo quality in these studies. In 34% (21/62) of the studies (5 TUNEL, 4 SCSA, 7 SCD, and 7 comet), a significant inverse relationship between sperm DNA fragmentation and embryo quality was reported, whereas the remaining 41 studies (18 TUNEL, 14 SCSA, 7 SCD, and 2 comet) showed no significant relationship between these parameters.

Embryo quality was assessed on day 2 or 3 in 48 studies (15 TUNEL, 16 SCSA, 9 SCD, and 8 comet) and on day 5 in 13 studies (5 TUNEL, 5 SCSA, 2 SCD, and 1 comet). Of the 62 studies, 36% of IVF studies (8/22), 21% of ICSI studies (5/24), and 50% of mixed IVF+ICSI studies (8/16) reported a significant correlation between sperm DNA damage and embryo quality (Table 23.3). When the studies were analyzed for embryo markers , delayed embryo development was associated with sperm DNA fragmentation in 37.5% of the studies (9/24), whereas an association between sperm DNA fragmentation and embryo grade was reported in 27.0% of studies (10/37). Studies involving the alkaline comet assay (78%) reported an adverse effect, whereas 22% of TUNEL, 22% of SCSA, and 42% of SCD reported adverse effect of sperm DNA fragmentation on embryo quality. A complete description of studies associating sperm DNA fragmentation with embryo quality is presented in Table 23.2.

Table 23.3 Summary of studies associating sperm DNA fragmentation with embryo quality

These results presented here are consistent with an earlier systematic review, reporting no clear relationship between sperm DNA damage and embryo quality [62]. Although experimental models (where there is diffuse and uniform sperm DNA fragmentation) show that sperm DNA fragmentation can have a profound effect on the developing IVF embryo [63, 64], this cannot be translated to human studies, where a wide spectrum of sperm DNA fragmentation is observed within an ejaculate [5, 65] and the quality of the sperm fertilizing the oocyte is unknown. This analysis reported a slightly higher proportion of the IVF studies (36%) and reported an association between sperm DNA damage and embryo quality than the ICSI studies (21%), although a statistical comparison is not possible due to the heterogeneity of these studies.

A higher proportion of the evaluable studies reported an association between sperm DNA fragmentation and embryo development (37.5%) than between sperm DNA fragmentation and embryo quality (27%). Embryos with a faster cleavage rate are more likely to develop into a blastocyst [66,67,68,69] and result in a successful pregnancy following transfer [70]. It is possible that extensive sperm DNA fragmentation could affect normal embryonic development [71, 72] by interfering with a variety of cellular processes, including DNA repair mechanisms, transcription, and cell cycle control [73, 74].

We observed a differential association between sperm DNA fragmentation and embryo quality, when the studies were segregated into groups based on the type of DNA fragmentation measurement assays. Sperm DNA fragmentation detected by the alkaline comet assay was associated with poor embryo quality in 78% of the studies compared to other assays. This association may be due to the increased sensitivity of the comet assays, where both single- and double-strand breaks are measured [75] and the intensity of broken DNA in the comet tail is directly proportional to the level of actual damage [76]. Although the TUNEL assay estimates DNA fragmentation directly, the unique organization of the sperm DNA [2] and the lack of a decondensation step to remove the protamines during the TUNEL protocol may result in the measurement of peripheral DNA damage rather than complete DNA fragmentation, thereby reducing the sensitivity of the assay. In conclusion, there is no consistent relationship between sperm DNA fragmentation and embryo quality (including embryo development); however the relationship between the two parameters may be associated with the ability of the assay to determine both single- and double-strand breaks.

2.5 Can Sperm DNA Fragmentation Predict ART Success ?

An extensive review of the existing literature and meta-analysis of studies testing the effect of DNA fragmentation on ART treatment was recently published by Simon et al. ([12], published online, ahead of print). In this meta-analysis (56 studies), clinical pregnancy was analyzed in 3734 IVF treatment cycles from 16 studies, 2282 ICSI treatment cycles from 24 studies, and 2052 mixed IVF+ICSI treatment cycles from 16 studies. An overall relationship between sperm DNA fragmentation and clinical pregnancy outcome from 56 studies (including 8068 ART cycles) supported a strong and significant association between the two parameters (odds ratio (OR) = 1.68; 95% CI, 1.49–1.89; P < 0.0001) [12].

The relationship between sperm DNA fragmentation and clinical pregnancy outcome were analyzed based on the type of treatment. A significant association between sperm DNA fragmentation and clinical pregnancy was observed for IVF treatment (OR = 1.65; 95% CI, 1.34–2.04; P < 0.0001), ICSI treatment (OR = 1.31; 95% CI, 1.08–1.59; P = 0.0068), and combined IVF+ICSI treatment (OR = 2.37; 95% CI, 1.89–2.97; P < 0.0001) [12]. The association between sperm DNA fragmentation and clinical pregnancy outcome was analyzed by assay type. The analysis with TUNEL assay studies (n = 2098 cycles from 18 studies; OR = 2.22; 95% CI, 1.61–3.05; P < 0.0001), SCD assay studies (n = 2359 cycles from 8 studies; OR = 1.98; 95% CI, 1.19–3.3; P = 0.0086), and alkaline comet assay studies (n = 798 cycles from 7 studies; OR = 3.56; 95% CI, 1.78–7.09; P = 0.0003) all showed a strong association between the two parameters. However, this was not the case with SCSA studies (n = 2813 cycles from 23 studies; OR = 1.22; 95% CI: 0.93–1.61; P = 0.1522) [12]. The meta-analysis was performed using the fixed and random effect models (Table 23.4).

Table 23.4 Odds ratio of sperm DNA fragmentation and clinical pregnancy outcome

Previous published meta-analysis [38, 77,78,79] has concluded that there is very little data to show a relationship between sperm DNA fragmentation and clinical pregnancy outcomes. However, the recent meta-analysis drawn across 56 eligible studies [12] reported a strong and a significant association between the two parameters. Overall, irrespective to the type of treatment or the method of analysis, the sensitivity and specificity of sperm DNA testing to predict clinical pregnancies were 32.6% and 76.4%, respectively, while the positive and negative predictive values to predict a clinical pregnancy were 71.2% and 38.8%, respectively. Although, we observe an increase in the positive and negative predictive values using the alkaline comet assay (81.7% and 46.8%, respectively), the overall predictive value suggests a moderate but significant association between sperm DNA fragmentation and clinical pregnancy.

In conclusion, the updated meta-analysis [12] reported a modest but significant association between sperm DNA damage and clinical pregnancy rate in all three ART treatment groups (IVF, ICSI, and mixed IVF+ICSI studies) with a variable effect according to the type of sperm DNA assay. One explanation for a moderate relationship between the two parameters may be due to patient inclusion factors. Most of the studies have included couples with female factors, and therefore the effect of sperm DNA damage on pregnancy outcome is compromised by female infertility factors [19]. Specifically, in the studies [56, 80,81,82] more than half of the couples had been diagnosed with female infertility. Whereas, in studies where patients with female infertile factor were eliminated, the odds to predicting a successful pregnancy have significantly increased irrespective to the type of DNA fragmentation testing method [19, 83]. In conclusion, the adverse effect of sperm DNA fragmentation on clinical pregnancies is observed in both IVF and ICSI treatments ; however, the odds to predict a clinical pregnancy may vary according to the type of assay used to measure sperm DNA fragmentation.

2.6 Is Sperm DNA Fragmentation Associated with Pregnancy Loss ?

The existing data associating sperm DNA fragmentation with spontaneous pregnancy loss is limited, yet a negative impact of DNA fragmentation on miscarriage following ARTs is observed in most studies. A meta-analysis [84] analyzed the association of these two parameters by identified 16 articles which included seven articles involving SCSA [14, 30, 34, 55, 85,86,87], six articles involving TUNEL assay [40, 41, 81, 88,89,90], two articles involving comet assay [61, 65], and one article involving acridine orange slide-based staining method [91]. These articles included 14 ICSI studies and 11 IVF studies comprising 2969 couples undergoing ART treatment resulting in 1252 pregnancies and 225 spontaneous pregnancy losses [84].

The results of the meta-analysis [84] suggested a significant increase in miscarriage in patients with high DNA damage compared with those with low DNA damage (Relative risk (RR), 2.16; 95% CI, 1.54–3.03; P < 0.0001). The meta-analysis also reported a strong association of DNA fragmentation measured by SCSA and TUNEL assays with miscarriages, while that of comet and acridine orange assays did not reach a statistical significance. The impact of DNA fragmentation on miscarriages was observed when DNA fragmentation was measured in the raw semen (RR, 1.65; 95% CI, 1.66–2.33; P < 0.0001) as well as the density gradient prepared subpopulation (RR, 3.47; 95% CI, 2.13–5.63; P < 0.0001; [84]). Earlier meta-analysis [92] also reported a positive impact of sperm DNA fragmentation on spontaneous pregnancy loss.

Aspects involving spontaneous pregnancy loss are not well understood. However, both maternal and paternal factors are known to be associated with pregnancy loss [93]. The negative impact of sperm DNA fragmentation is more pronounced in animal models where this leads to abnormal embryo development, reduced implantation rate, and frequent pregnancy loss [63, 64, 94]. Such prolonged effect of sperm DNA fragmentation, also known as the late paternal effect [95], may be in part due to the inability of the oocyte to repair the damaged sperm chromatin when it exceeds the threshold value [72]. In conclusion, the findings of the recent meta-analysis suggest that sperm DNA fragmentation is positively associated with pregnancy loss after IVF and ICSI treatments [84].

3 Clinical Value of Tests of Sperm DNA Fragmentation

The evidence based on the current literature search supports the fact that there is an influence of sperm DNA fragmentation on male reproductive health, and sperm DNA testing could be incorporated into routine clinical use [96], although some clinical reviews and meta-analyses do not support the clinical use of sperm DNA fragmentation [78, 97]. Despite controversies, in recent years there has been a marked increase in the commercial sperm DNA testing for clinical use. In the following sections, we will discuss some of the evidence in support of the use of sperm DNA testing in infertile men.

3.1 Sperm DNA Fragmentation as a Biomarker

DNA fragmentation is a common property of the sperm. Once the damage occurs, the sperm lacks any mechanism to fix it, and therefore the fragmentation occurring to sperm chromatin is a permanent change. Sperm DNA fragmentations is showed to be higher in patients with infertility issues [13,14,15,16,17,18,19,20,21] and associated with abnormal semen parameters [33, 40,41,42, 44, 52, 55, 56, 86]. Studies on time to pregnancy have suggested that sperm DNA fragmentation is an excellent predictor of natural conception [14, 28]. Sperm DNA fragmentation is associated with advanced male age [34], exposure to environmental toxins [98, 99], in cancer patients and treatments [65, 100,101,102], and infertility conditions such as varicocele [18, 103,104,105]. Sperm DNA fragmentation is also a useful biomarker for various end points during ARTs, such as fertilization rate [41, 43, 106, 107], embryo quality [44, 87, 108], embryo development [41, 65, 109], clinical pregnancy [81, 82, 106, 107], miscarriage [41, 55, 81], and live birth [61, 81, 110]. In conclusion, sperm DNA fragmentation can be used as a biomarker independent of semen parameters.

3.2 Diagnosis of Male Infertility

Male factor infertility is the primary cause of infertility in approximately 20–30% of infertile couples. Male factor infertility is contributing factor in another 30–40% of infertile couples in addition to female factors [111, 112]. Thus, male factor infertility is present in half of all couples with infertility issues. To date, the routine method of male factor infertility diagnosis is based on traditional semen analysis [113]. Recently, it is estimated that approximately 15–20% of men with infertility issues have a normal semen analysis profile [114], and therefore a definitive diagnosis of male infertility cannot be performed only on the basis of semen analysis. A number of studies comparing sperm DNA fragmentation status between fertile and infertile men [13,14,15,16,17,18, 20, 21, 115] have suggested that DNA fragmentation could be a useful biomarker for male infertility diagnosis. Although, sperm DNA fragmentation is correlated with abnormal semen parameters such as sperm concentration, motility, and morphology [33, 40, 41, 44, 52, 55, 56, 86, 116], this is not a consistent finding in all studies [65, 81, 88, 91, 117,118,119]. Therefore, sperm DNA fragmentation can be considered an independent factor of male infertility and should be used together with the semen analysis as an added parameter to diagnose the status of male reproductive health.

3.3 Counseling Infertile Couples Prior to Initiating Infertility Treatment

As discussed earlier, the structure and stability of the sperm chromatin are important for male fertility and normal in vitro fertilization. In a study by Evenson et al. [14], using the sperm chromatin structure assay, DNA fragmentation above the threshold value of 30% is associated with a significant reduction in male infertility potential, and the time to pregnancy in vivo is longer in patients with DNA fragmentation >30% compared to <20%. The likelihood of achieving a natural pregnancy is significantly lower when DNA fragmentation is greater than 30% [120]. Another study examining the effect of DNA fragmentation among first-pregnancy planners with no previous knowledge of their fertility capability suggested that fecundability declined with an increase in sperm DNA fragmentation [28]. These reports summarize that when DNA fragmentation exceeds the threshold value , then male fertility is significantly reduced [14, 28]. Therefore, in couples who are planning for first pregnancy, the DNA fragmentation assay is a good predictor of male fertility potential and negative pregnancy outcomes. In couples where the male partner has extensive DNA fragmentation, counseling to improve their reproductive health and strategies to reduce the level of sperm DNA fragmentation should be provided (Grade B recommendation).

Sperm DNA testing can also be useful in another area of pretreatment counseling involving IUI treatment. A meta-analysis performed in this chapter (unpublished data) involving all the available studies using SCSA [18, 30,31,32, 34] reported a positive predictive value of 96% when DNA fragmentation is above the threshold (30%), while the sensitivity of the assay was 16%. In couples with high sperm DNA fragmentation, assisted reproduction (IVF or ICSI) should be considered.

3.4 Counseling Infertile Couples Planning to Choose Assisted Treatment

The question still remains whether ICSI would be a beneficial treatment of choice if the male partner presents with an increased level of sperm DNA fragmentation. An increase in success rate following ICSI [121, 122] may be attributed to a lower proportion of female factor infertility, where improved oocyte quality would be associated with a better DNA repair capability of the oocytes when fertilized with DNA fragmented sperm [123]. It can also be postulated that selection of physiologically motile and morphologically normal sperm for ICSI insemination [50] by the embryologists increases the probability of choosing sperm with low DNA fragmentation. In such conditions, the probability of selecting sperm with relatively low DNA fragmentation is higher compared to the overall sperm population as these factors (normal morphology and progressive motility) are inversely correlated with sperm DNA fragmentation [19, 33, 40,41,42, 44, 52,53,54, 124].

We performed a literature search to identify studies that simultaneously performed sperm DNA fragmentation assays (SCSA, TUNEL, SCD, and comet assays) on patients undergoing IVF and ICSI inseminations . We identified 23 studies that fit this criterion, following elimination of studies using mixed (IVF and ICSI combined) patient group. Of the 23 studies, 18 studies compared the effect of sperm DNA fragmentation with clinical pregnancy outcome following IVF and ICSI treatments. The relationship between sperm DNA fragmentation (above and below the threshold value) and clinical pregnancies (after IVF vs. ICSI insemination) was analyzed by two-by-two table obtained from 15 studies [19, 30, 31, 34, 40, 41, 54, 55, 61, 85, 105, 119, 125,126,127], while data was not available for three studies [128,129,130]. This analysis included 5564 treatment cycles from 3853 IVF cycles and 1711 ICSI cycles (unpublished data). Pregnancy rates were comparable between IVF and ICSI treatment when DNA fragmentation was below the threshold value, 34.19% for IVF and 37.15% for ICSI treatment (Chi Sq. = 2.847; df = 1; p = 0.0915). However, when the clinical pregnancies were analyzed when sperm DNA fragmentation was above the threshold value, then ICSI had a higher clinical pregnancy rate (32.14%) compared to IVF (16.41%) treatment (Chi Sq. = 20.815; df = 1; p < 0.0001). As expected the clinical pregnancy rates were higher in both IVF and ICSI treatments when sperm DNA fragmentation is below the threshold value (unpublished data).

A comprehensive large study by Simon et al. [72], comparing the quality of 2210 embryos (observed on day 2, 3, and 5) obtained from IVF and ICSI insemination at different levels of sperm DNA fragmentation, reported that the quality of ICSI embryos are significantly higher than IVF embryos when patients are presented with high sperm DNA fragmentation (Table 23.5). The literature presented above provides sufficient evidence to show that ICSI treatment is not affected by the level of sperm DNA fragmentation. In fact, the ICSI treatment resulted in twice the amount of pregnancies compared to IVF treatment when the DNA fragmentation was above the threshold value. A meta-analysis reported no difference in miscarriage rates following IVF or ICSI treatments when sperm DNA fragmentation is above the threshold value [84]. It can be argued that the outcomes following ICSI treatment may depend on the oocyte quality, where the negative impact of high DNA fragmentation on ART outcomes can be overcome by good quality oocytes that are able to repair the fragmented sperm DNA [131]. Since the success of ARTs has great emotional, financial, and age-related consequences for the couple, the selection of an appropriate treatment (IVF or ICSI) may favor the patient’s success.

Table 23.5 Comparison of embryo quality between IVF and ICSI insemination methods at three levels of sperm DNA fragmentation

Based on the results of our meta-analysis [12], the data suggest that tests of sperm DNA damage may provide some predictive value in the context of IVF, ICSI, and mixed IVF + ICSI. An analysis of the 16 IVF studies (with a median pregnancy rate of 32%) revealed a median PPV of 79% and median NPV of 35%. This means that in populations with an overall IVF pregnancy rate of 32%, sperm DNA tests can discriminate between IVF pregnancy rates of 21% (positive test) and 35% (negative test), which represent a clinically important difference in pregnancy rate. An analysis of the 24 ICSI studies (with a median pregnancy rate of 36%) revealed a median PPV of 64% and median NPV of 40%. In the context of ICSI, sperm DNA tests can discriminate between ICSI pregnancy rates of 36% (positive test) and 40% (negative test), which is a small difference of modest clinical value. With the 16 mixed (IVF + ICSI) studies, we observed a median PPV of 70% and median NPV of 50%, suggesting that in populations with an overall mixed (IVF + ICSI) pregnancy rate of 44%, sperm DNA damage assessment can discriminate between mixed (IVF + ICSI) pregnancy rates of 30% (positive test) and 50% (negative test), a notable difference in pregnancy rate of important clinical value. Therefore, couples with high sperm DNA fragmentation and enrolled in an IVF treatment cycle should proceed to ICSI rather than IVF (Grade C recommendation).

3.5 As a Biomarker for Reproductive Toxicological Studies

Sperm are particularly vulnerable to xenobiotic action which can result in DNA damage [132]. The exposure to xenobiotics can be classified into three major types such as occupational exposure, environmental exposure, and pharmacological exposure. Studies have shown that sperm DNA fragmentation is higher among coke oven workers in contact with polycyclic aromatic hydrocarbon exposure [133]. Oh et al. [134] had shown that there are elevated levels of DNA fragmentation among the waste incineration workers when compared with men from similar origin. Men working in the factories in contact with organic molecules such as styrene show a significant amount of increase in sperm DNA fragmentation [135]. Similarly, men working in the insecticide and pesticide industries have been proven to show increases in sperm DNA fragmentation [136, 137]. Other occupational exposure such as farmers exposed to insecticides is also known to significantly stimulate sperm DNA fragmentation [136, 137]. Workers exposed to organic chemicals are also reported to exhibit high levels of sperm DNA fragmentation [135].

Today, pharmacological exposure has become very common due to advances in molecular medicine, especially in the field of cancer. Pharmacological intervention for the treatment of diseases results in genotoxicity to sperm and male germ cells. Such exposures are genotoxic to the male germ cells and cannot be avoided. A well-known example for such intervention is cyclophosphamide , which is used as chemotherapeutic agents to treat cancer [138]. Environmental exposure to xenobiotics cannot be avoided in the present-day life because these pollutants are present along with food or water or air. Environmental estrogens and similar compounds are known for their effect on male infertility and sperm DNA fragmentation [139]. Some of the other environmental pollutions that have the ability to induce DNA fragmentation are organo-chlorides [140], smog [98]. Aitken et al. [141] suggested that paternal exposure to xenobiotics not only results in genetic or epigenetic changes to the sperm but also causes adverse consequences for the offspring. Exposure to xenobiotics can result in high levels of sperm DNA fragmentation beyond the capacity of the oocyte to repair and can result in preimplantation failure [132]. Number of studies support the concept that exposure to xenobiotics can have a powerful impact on sperm DNA and its function. Therefore, sperm DNA testing may not only be useful to identify male reproductive health status but also is a method commonly used for toxicological studies (Grade C recommendation).

3.6 Unexplained Infertility

We know that 25–30% of couples undergoing ARTs are diagnosed with unexplained infertility [142]. In these cases, men have no obvious history of fertility problems and physical conditions or endocrine issues, and the semen analysis results are normal [143]. The prevalence of high DNA fragmentation is showed in men with unexplained infertility [60, 144, 145]. In a study, when 147 unexplained infertile men were screened for sperm DNA fragmentation, 84% of these patients had DNA fragmentation above the 25% cutoff value used to determine fertile from infertile men [61]. Further analysis of the study [61] reported that approximately 41% of men categorized with unexplained infertility issues have sperm DNA fragmentation above the threshold (52%) to obtain a clinical pregnancy following IVF treatment [61]. Similarly, another study using the SCSA reported that 26% of men diagnosed with unexplained infertility had high DNA fragmentation index [60]. Feijo et al. [145] reported that men with unexplained infertility have high level of sperm DNA fragmentation measured by SCD and TUNEL assays . More than 60% of men with unexplained infertility are showed to have abnormal protamine profile [19], which could be a causative agent for increased sperm DNA fragmentation [20]. These results suggest that to some extent sperm DNA fragmentation assays may help to identify men with fertility problems even when they are presented with normal semen analysis, as reported in unexplained infertility cases. Therefore, in men with unexplained infertility sperm, DNA testing can be used as additional marker of sperm quality to help in the counseling of these couples (Grade C recommendation).

4 Why Sperm DNA Testing Is Not Routinely Used Clinically? “Sperm DNA Testing: Pro” Point of View

Sperm DNA test has been performed for more than 30 years. However, vast majority of the data associating the effect of sperm DNA fragmentation on clinical outcomes is published in the last 20 years. A recent literature search using terms related to “sperm DNA damage ,” “sperm DNA fragmentation,” and “sperm DNA integrity ,” along with “male infertility,” “ART,” “IVF,” and “ICSI” yielded more than 1300 related articles. Despite a well-studied area of research, controversies do exist as to the effect of sperm DNA fragmentation on male infertility and assisted reproductive outcomes. The controversy is largely due to the fair quality of the available studies. In this section, we highlight few facts that would help the readers understand the discrepancies regarding the use of sperm DNA fragmentation for clinical use.

4.1 Sperm DNA Testing Is a Broad Term Used to Refer to a Number of Assay Methods

There are four widely used methods to access sperm DNA fragmentation: the comet assay [13], TUNEL assay [146], SCSA [14], and SCD assay [147]. The comet and TUNEL assays are the detect methods to measure DNA strand breaks, while the SCSA and SCD indirectly measure chromatin integrity by measuring the susceptibility of DNA to denaturation [97]. These assays are known to measure different aspects of sperm DNA fragmentation [148, 149], while the ability of these assays to accurately measure the level of DNA fragmentation depends on the technical and biological aspects of each test [150]. A recent meta-analysis [12] associating the four DNA fragmentation assays with ART outcomes suggests that the prognostic value of these assays is different [12]. The odds ratio (OR, 2.35; 95% CI, 1.99–2.78; P < 0.001) and relative risk (RR, 1.35; 95% CI, 1.28–1.43; P < 0.001) to predict a clinical pregnancy by the direct methods (TUNEL and comet assays including 2897 ART cycles) were higher than the indirect methods (SCSA and SCD assays including 5172 ART cycles) odds ratio (OR, 1.12; 95% CI, 0.98–1.28; P = 0.096) and relative risk (RR, 1.04; 95% CI, 0.99–1.09; P = 0.089). Although, DNA fragmentation measured by the indirect methods were not significantly associated with clinical pregnancy following ARTs (P > 0.05), the overall predictive value (obtained by combining all four sperm DNA tests) was significant [12]. The fact that prior meta-analyses [38, 77,78,79, 92] have reported an uncertain effect of sperm DNA fragmentation on ART outcome is likely due to the heterogeneity of the studies (combining studies using direct and indirect methods). We believe that future meta-analysis should be performed with independent analyses based on assay type.

4.2 The Structural Organization of the Sperm Chromatin Is an “Achilles Heel ”

As discussed earlier, each assay measures different aspects of sperm DNA fragmentation. The available literature suggests that there is a wide difference in the threshold values between the assays, not referring to the variations in threshold values observed within the assays. A study comparing three different assays on the same patient population suggests a difference in threshold value, the comet assay (82%), TUNEL (10%), and SCSA (27%; 75), while the threshold for SCD assay may vary between 30% and 35% [44, 105]. These threshold values are primarily based on the range of sperm presented with DNA fragmentation measured by each assays. For example, the ability of an assay to determine the level or percentage of sperm with DNA fragmentation within an individual is variable: TUNEL (0–66%), SCSA (0–44%), SCD assay (0–50%), and the alkaline comet assay (0–100%) [37, 55, 75, 125, 126, 151].

Are the observed differences between assays related to the ability of these assays to access the sperm chromatin? In other words, how much sperm chromatin is accessible within the sperm head by these DNA fragmentation assays? To answer this question, we first have to understand the structural organization of sperm chromatin [152, 153]. The sperm head consists of one half of the genome and tightly packed with the help of protamines [2]. It is reported that the sperm chromatin is six times more compact than that of somatic cells and it is almost crystalline in nature [154]. Recently, Simon et al. [5] demonstrated that the volume of sperm nuclei (28.2 ± 0.2 μ[mu]m3, [155]) is almost doubled (~63 μ[mu]m3) by the process of decondensation within the intact sperm (without breaking the cell wall), and the process of lysis and decondensation results in 34-fold increase in volume (~1018 μ[mu]m3) of the sperm chromatin (after breaking the cell wall). Another experiment [156] reported that decondensation of sperm nucleus increases the ability of TUNEL to detect increased levels of DNA fragmentation in sperm. Similarly, during the alkaline comet assay, the reduction of disulfide bonds connecting protamines using DTT [157] and removal of protamines under alkali conditions [158] help to relax the sperm chromatin [5] and access the complete level of DNA fragmentation within the sperm. In conclusion, due to the condensed nature of the sperm chromatin, the sperm DNA assays should consider decondensation of sperm chromatin prior to DNA fragmentation analysis.

4.3 Infertility Is a Multifactorial Issue, While Testing Sperm DNA Testing Is One Part of an Equation

For the past 30 years, several studies have identified factors that could influence the success of an assisted treatment. These factors can be broadly classified into female-derived factors, male-derived factors, and embryonic factors (prior to transfer). In a meta-analysis involving commonly reported factors that are known to influence ART success was reported by van Loendersloot et al. [159]. This meta-analysis reported that some factors (female age, duration of subfertility, basal FSH level) were negatively associated with ART success, and the number of oocytes retrieved was positively associated with ART success , while other factors (parity, indication for subfertility, method of fertilization, and number of embryos transferred and embryo development) was not significantly associated with ART success [159]. It is reasonably known that transfer of good quality blastocysts has increased chances of pregnancy than lower-quality embryos, early-stage blastocysts, or cavitating morula [160, 161]. Among the male-derived factors, male age and functional quality of sperm had no influence on ART success [162]. Based on the recent meta-analysis [12], we now know that there is an association between sperm DNA fragmentation and ART outcome. Despite a significant association between the two parameters, the predictive value of sperm DNA fragmentation to achieve a successful pregnancy is low [19, 75]. One explanation for the low predictive value of sperm DNA fragmentation on ART success can be attributed to the involvement of female-derived factors, where in couples with female factors, the effect of sperm DNA fragmentation on pregnancy outcome is compromised. In support of this theory, a study by Simon et al. [19] concluded that sperm DNA fragmentation can be an independent factor to predict a successful pregnancy (OR, 76.00; 95% CI, 8.69–1714.44; P < 0.001) following elimination of couples with known female factors and cases with unexplained causes of infertility. In conclusion, the effect of sperm DNA fragmentation on ART success is likely diminished by the presence of female factors . Therefore, future studies associating these two parameters (sperm DNA fragmentation on ART success) should control for female factors to accurately determine the clinical value of sperm DNA fragmentation [83].

4.4 Reliable Testing and Reporting of Sperm DNA Fragmentation

Four assays are commonly used to determine the level of DNA fragmentation in sperm. The current literature reports a wide range of variations within each assay in terms of assay protocol, software used for analysis, the type of sample used (fresh or frozen), study population and control, and reporting of results. These variations observed within the studies question the reliability of DNA fragmentation testing. Assays having standardized protocol along with automated software for analysis such as the SCSA [14] have a strict threshold value ranging from 27% to 30% [86, 163, 117]. When the specified software [14] for SCSA is not used, the ability of the assay to predict ART success at the specified threshold (27–30%) value is not observed [55, 80, 127, 164], and such discrepancies are understandable, while some research groups using alternative software for SCSA have established their own threshold value outside the specified range 27–30%. These studies report a significant association between SCSA and ART outcomes at a lower threshold value [32, 165]. When all the studies using SCSA are summarized using a meta-analysis, the association between DNA fragmentation and ART success is reduced [12].

The slide-based TUNEL assay has a standard protocol (using the commercial kit) but does not use a software to compute the results (the reporting of results is solely based on the technician’s observation). In this case, a wide range of threshold value has been reported, 4% [54, 125], 10% [40, 75], 15% [41, 90], 17.6% [108], 20% [52, 88], 35% [81], 36.5% [53], and 48% [118]. Here we see the authors have established the threshold values according to their laboratory conditions, which means the suggested threshold value may not work at another laboratory setting or a different technician. Despite such wide range of threshold values established for TUNEL assay (4–48%), the meta-analysis suggests a strong correlation between DNA fragmentation and ART success [12]. In conclusion, we rely on assays that have been validated with testing of control-fertile populations and with well-established thresholds. Variation in the threshold values may raise concerns, but in cases when there is no standard software to compute the results, threshold values should be established according to the specific laboratory settings.

4.5 Association Between Sperm DNA Fragmentation and ART Outcomes Is a Subject of Experimental Bias

It is well-known that sperm DNA fragmentation may vary with time. Within an ejaculate, the level of DNA fragmentation is not consistent in all the sperm; some sperm are vulnerable to DNA fragmentation and some don’t. In a recent study using the comet assay , it is showed that the level of DNA fragmentation in the sperm population may vary and could be potentially classified into three types [5]. Therefore, under natural conception the sperm fertilizing an oocyte is a random event (in terms of DNA fragmentation) as motile sperm are known to carry fragmented DNA [61, 119]. In accordance with the random selection effect, we see that in all studies conducted on DNA fragmentation and time to pregnancy with first-pregnancy planners, the time for conception increases with an increase in sperm DNA fragmentation [14, 28, 166]. A meta-analysis (presented in Sect. 2.2) consisting of 1135 IUI cycles suggested that sperm DNA fragmentation is strongly associated with IUI outcome, although selection of a motile sperm population by density gradient centrifugation (DGC) is subjected to experimental bias unless DNA fragmentation analysis is performed in the prepared sperm population. Under natural conception as well as IUI treatment, the sperm fertilizing an oocyte is selected randomly in terms of sperm DNA fragmentation. This is not surprising because we know that randomization reduces selection bias and allows us to determine any effects of the treatment efficiently.

Let’s consider the scenario of studies that associate sperm DNA fragmentation with IVF or ICSI outcomes. Experimentally, the first line of bias arises from the sperm selection process . Most often in assisted reproduction, DGC is the standard procedure to select the prepared sperm population. Studies comparing DNA fragmentation before and after DGC have reported that the level of DNA fragmentation is reduced in the prepared sperm population [5, 119]. In conventional IVF treatment, the sperm fertilizing an oocyte is random (although prepared sperm population is used), while during ICSI treatment, the embryologists further select for physiologically motile and morphologically normal sperm. Since DNA fragmentation is negatively associated with progressive motility and normal morphology [19, 40, 42, 54, 124], the second sperm selection process may facilitate an additional selection of sperm with absent or low of DNA fragmentation. This is in support of the results presented in the meta-analysis (presented in Sect. 3.4), where the clinical pregnancy rates are twice higher after ICSI treatment compared to IVF treatment when DNA fragmentation is above the threshold value.

The second line of experimental bias arises at the embryo transfer stage . It is well documented that a successful pregnancy is favored by the transfer of high-quality embryos [70]. Irrespective to IVF or ICSI treatments, the best quality blastocysts (generally two) are selected for transfer from the pool of available embryos. Using animal model [63] and in human subjects [72, 167], sperm DNA fragmentation is reported to influence the quality of the embryo. Therefore, in both IVF and ICSI treatments, the good quality embryo used for transfer is presumably fertilized by reasonably good quality sperm, and such sperm does not represent the initial level of DNA fragmentation measured in the ejaculate. Here we see the process of randomization is absent, and during the process of assisted treatment, randomization is impossible.

In studies that associate sperm DNA fragmentation with clinical pregnancy outcome , we assume that the sperm used to fertilize the embryos that were subsequently transferred represents the population of sperm in the ejaculate (in terms of DNA fragmentation). However, this is subjected to experimental bias as these sperm were not randomly selected but in fact selected at two stages. Despite a biased experimental design, several studies and meta-analyses have demonstrated significant association between sperm DNA fragmentation and pregnancy outcome is simply astonishing [12]. We suspect that the effect of sperm DNA fragmentation on ART outcome would be greater if each step during the process of ART was randomized, but such a scenario is not possible under the current clinical setting.

The third line of experimental bias arises in assisted reproduction when the couple undergoing assisted treatment is presented with female factor infertility . Very few studies in the available literature have controlled for female infertility factors and those who did reported a highly significant association between sperm DNA fragmentation and ART outcome [12]. As discussed in Sect. 4.3, despite absence of DNA fragmentation, the presence of female infertility factors may reduce the chances of a successful pregnancy. In a clinical setting of couples where half of them are presented with female infertility factor, the probability of a pregnancy is affected by 50%, even if the sperm have normal DNA. In conclusion, we need to recognize that assisted reproduction is heavily biased. Despite such bias, the current literature [12] still shows a significant negative association between sperm DNA fragmentation and ART outcome supporting a true relationship between the two parameters.

5 Conclusion

A clinically useful test for sperm function should have the following characteristics : reliability, cost-efficiency, ability to predict outcomes in vitro and in vivo, and ability to assist clinicians in decision-making [168, 169]. Conventional semen analysis is the only test currently available for men with infertility issues. Although sperm DNA fragmentation shows a correlation with sperm parameters (especially sperm motility, viability, and morphology), these correlations are generally weak. Comparative studies of infertile men and fertile controls report a strong association between sperm DNA fragmentation and infertility. Therefore, sperm DNA fragmentation is associated with male infertility independent of semen parameters. Studies associating semen parameters and fertility in populations of first-pregnancy planners report a weak or no association between the two parameters [170,171,172]. In contrast, a strong association is observed between sperm DNA fragmentation and time to pregnancy [14, 28] suggesting the importance of sperm DNA integrity testing for male reproductive health.

The primary function of the sperm is to deliver the haploid genetic material to the oocyte . During the process of assisted reproduction, sperm is delivered near the oocyte (IVF) or delivered into the oocyte (ICSI). In such case, it is not reasonable to expect any of the semen parameters to predict ART success [119]. Presumably, sperm DNA fragmentation should be more closely associated with ART outcomes. The meta-analyses and systematic reviews presented here demonstrate that sperm DNA fragmentation is a good predictor of IUI failure and is associated with IVF pregnancy but less so with ICSI outcomes. Sperm DNA fragmentation is also negatively associated with embryo development and implantation and positively associated with miscarriage rates [12].

Another clinical utility of sperm DNA fragmentation is to assist clinicians and infertile couples to help choose the type of assisted treatment. The meta-analysis review presented here on IUI cycles suggests that a sperm DNA fragmentation above the threshold value results in a 96% IUI failure rate. The meta-analysis on IVF and ICSI cycles demonstrates that in couples with DNA fragmentation above the threshold value, clinical pregnancies following ICSI (32.14%) are double than following IVF (16.41%) treatment.

Overall, the evidence presented here (both in vivo and in vitro) and elsewhere favors sperm DNA fragmentation testing as a potential and clinically useful biomarker to predict male infertility and ART outcome [173, 174]. We support the view that sperm DNA testing should be done in addition to the conventional semen analysis for a complete diagnosis of male reproductive health.