Are sperm chromatin and DNA defects relevant in the clinic?

Abstract

There has been an increase in the use of sperm DNA and chromatin integrity tests in the evaluation of the infertile man with the hypothesis that these tests may better diagnose infertility and predict reproductive outcomes. This review discusses the etiology of sperm DNA damage, briefly describing the tests of sperm DNA damage, and evaluates the relationship between sperm DNA damage and reproductive outcomes. A systematic review of the literature allows us to conclude that sperm DNA damage is associated with lower natural, intra-uterine insemination (IUI), and in vitro fertilization (IVF) pregnancy rates. Studies to date have not shown a clear association between sperm DNA and chromatin defects and pregnancy outcomes after intra-cytoplasmic sperm injection (ICSI). However, we cannot exclude the possibility that very high levels of DNA damage will impact on ICSI outcomes. In couples undergoing IVF or ICSI, there is evidence to show that sperm DNA damage is associated with an increased risk of pregnancy loss. A limitation of this systematic review and meta-analysis is that it does not address the heterogeneity of the individual study characteristics. Although the clinical utility of tests of sperm DNA damage remains to be firmly established, the data suggest that there is clinical value in testing couples prior to assisted reproductive technologies (ARTs IUI, IVF, and ICSI) and in those couples with recurrent miscarriages. Additional, well-designed prospective studies are needed before testing becomes a routine part of patient care.

Keywords:

• insemination
• male infertility
• miscarriage
• pregnancy
• sperm DNA

Introduction

Standard semen parameters exhibit a high degree of biological variability. They are only fair measures of fertility potential and poor predictors of reproductive outcomes [Guzick et al. 2001]. As such, there is a need to develop new markers that differentiate between infertile and fertile men, and help predict pregnancy outcome and adverse reproductive events. Tests of sperm chromatin and DNA integrity can help in the diagnosis of male infertility [Evenson et al. 1999; Kodama et al. 1997; Spano et al. 2000; Zini et al. 2001]. Moreover, it has been reported that tests of sperm chromatin and DNA damage may help predict reproductive outcomes after assisted reproductive technologies (ARTs) but the value of these tests in the clinic remains to be fully delineated [Bungum et al. 2007].

ARTs have revolutionized the treatment of male infertility. With in vitro fertilization (IVF) with intra-cytoplasmic sperm injection (ICSI), men with severe oligozoospermia, obstructive and non-obstructive azoospermia can hope to father children. However, some questions regarding the safety of ARTs remain. These safety concerns are relevant because: (i) ARTs can circumvent the hurtles of natural selection, (ii) infertile men with severe male-factor possess substantially more sperm DNA damage than fertile men, (iii) pregnancy is possible regardless of the degree of sperm DNA damage, and (iv) experimentally, fetal and post-natal development is related to the degree of sperm DNA damage [Evenson et al. 1999; Fernandez-Gonzalez et al. 2008; Gandini et al. 2004; Kodama et al. 1997; Spano et al. 2000; Zini et al. 2001].

Etiology of sperm DNA damage

The etiology of sperm DNA damage is multi-factorial. Clinically, several conditions including chemotherapy, smoking, genital tract infection, and varicocele have been associated with sperm DNA and chromatin damage [Banks et al. 2005; Bungum et al. 2007; Erenpreiss et al. 2002; Fossa et al. 1997; O'Flaherty et al. 2008; Potts et al. 1999; Sailer et al. 1997; Saleh et al. 2003; Zini and Sigman 2009]. Broadly, these conditions can be categorized as primary or intrinsic defects in spermatogenesis, i.e., genetic or developmental abnormalities and secondary or extrinsic noxious factors, i.e., gonadotoxins, hyperthermia, oxidants, or endocrine disruption.

At the cellular level, a number of theories have been proposed to explain the DNA damage in human spermatozoa. Studies have suggested that protamine deficiency (with aberrant chromatin remodeling), reactive oxygen species (ROS), abortive apoptosis, and alterations in topoisomerase II activity may be responsible for sperm DNA damage [Aitken and De Iuliis 2007; Aoki et al. 2006; 2005; Cho et al. 2001; Leduc et al. 2008; Sakkas et al. 2003; Tarozzi et al. 2007]. Recently, De Iuliis et al. [2009], proposed a two-step hypothesis to explain the generation of sperm DNA damage. Based on this work, oxidative stress acts on poorly protaminated cells, i.e., cells with incomplete replacement of histones by protamines, generated as a result of defective spermiogenesis [De Iuliis et al. 2009].

Tests of sperm DNA damage

Several tests of sperm DNA and chromatin damage have been described [Chohan et al. 2006; Evenson et al. 1999; Zini and Sigman 2009]. These tests have been developed in the hope that they may: (i) help in the diagnosis of male infertility, (ii) predict reproductive outcomes in the context of ARTs, and (iii) provide some assurance regarding the integrity of the male gamete genome. Several factors must be considered when evaluating studies of sperm DNA and chromatin integrity. First, different assays measure different aspects of sperm DNA and chromatin, e.g., the degree of DNA fragmentation, level of protamination, and/or DNA denaturation. Second, the assay conditions can greatly influence the accessibility of the dye or enzyme to the sites of damaged DNA and, therefore, impact on the final results. For example, the concentration of reducing agents such as, dithiotreitol (DTT) can alter the degree of sperm nuclear decondensation. Third, assays are restricted because they do not clinically resolve the extent of DNA fragmentation or gene-specificity away from a normal threshold of DNA damage in a given cell. Fourth and finally, sample preparation and handling (e.g., centrifugation, prolonged incubation) prior to assessing sperm DNA and chromatin integrity can impact the final test results.

The COMET (single cell gel electrophoresis) and TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling) assays are commonly used to detect DNA strand breaks directly. Some assays measure the rate at which denatured single stranded DNA form from native double stranded DNA (e.g., SCSA-sperm chromatin structure assay). This depends on the premise that nicked DNA will denature more readily than intact DNA. Other assays rely on the differential binding of dyes or agents to single stranded and double stranded DNA (e.g., acridine orange) or to protamine-deficient sites (e.g., aniline blue or CMA3 test) [Chohan et al. 2006]. Remarkably, the results of most sperm DNA or chromatin integrity assays correlate. The exception is the microscopic acridine orange test [Chohan et al. 2006]. To provide clinically relevant information, an upper normal level (cutoff) of the percentage of cells with DNA fragmentation or chromatin defect has been set in most published studies. However, the method of defining the cutoff varies greatly (e.g., receiver operating characteristics, fertile populations, and population mean). Samples with test results above the threshold or cutoff value are considered to have high levels of DNA damage [Zini and Sigman 2009].

Relationship between sperm DNA damage and pregnancy

In order to assess the relationship between sperm chromatin/DNA damage and pregnancy outcomes a systematic review and meta-analysis of valid published studies was conducted. On one hand, the strength of systematic reviews (with meta-analysis) is the improved precision of the summary estimates compared with the individual study estimates of the relationship between sperm DNA/chromatin defects and pregnancy outcomes. On the other hand, a weakness of meta-analyses is that it combines studies with highly variable study characteristics. These characteristics include data collection (prospective or retrospective), population characteristics (unselected, male factor), female inclusion/exclusion criteria, laboratory expertise in assessment of sperm DNA/chromatin damage, sperm DNA/chromatin test type, and sperm DNA test cutoff.

Natural Pregnancy and Intra-uterine Insemination (IUI) Studies

Although small in number, the available studies suggest that sperm DNA and chromatin damage is related to lower rates of natural and IUI pregnancies. Studies have shown that sperm DNA damage is associated with a prolonged time to pregnancy and a very low probability of achieving a natural pregnancy [Evenson et al. 1999; Giwercman et al. 2010; Loft et al. 2003; Spano et al. 2000]. A systematic review and meta-analysis of published natural pregnancy studies clearly correlates sperm DNA damage and failure to achieve a natural pregnancy (odds ratio (OR) = 7.01, 95% CI 3.68, 13.36, p < 0.001; Table 1). Sperm DNA damage has also been associated with lower IUI pregnancy rates [Bungum et al. 2007; Duran et al. 2002; Muriel et al. 2006]. The Bungum et al. [2007] study permits one to construct a 2×2 table unlike the data in the Duran et al. [2002] and Muriel et al. [2006] studies. In the former, the diagnostic OR = 9.9 (95% CI 2.37, 41.51, p < 0.001) indicating that sperm DNA damage is related to a significantly lower pregnancy rate after IUI.

Table 1.  Selected Diagnostic Properties of Studies on Sperm DNA Damage and Natural Pregnancy.

Study	n	Assay	%hDD	Sens	Spec	PPV	NPV	OR	(95% CI)
Evenson et al. 1999	144	SCSA	7	0.19	0.96	0.60	0.81	6.54	(1.72, 24.92)
Spano et al. 2000	215	SCSA	13	0.23	0.96	0.86	0.55	7.59	(2.54, 22.67)
Giwercman et al. 2010	257	SCSA	12	0.21	0.96	0.83	0.58	6.82	(2.52, 18.47)

Abbreviations: %hDD = proportion of samples with high sperm DNA damage; Sens = sensitivity; Spec = specificity; PPV = positive predictive value; NPV = negative predictive value; OR = odds ratio; SCSA = sperm chromatin structure assay.

In Vitro Fertilization (IVF) Studies

Over the past 10 years, more than twenty studies have examined the relationship between sperm DNA integrity and pregnancy rates after standard IVF. It is important to note that these studies are quite heterogeneous in terms of their design (prospective/retrospective), inclusion/exclusion criteria (e.g., female age, female infertility factors), and types of sperm DNA test employed (SCSA, TUNEL). With this constraint, a systematic review and meta-analysis of the eleven IVF studies that can be evaluated by a 2 × 2 table shows that sperm DNA damage is associated with lower IVF pregnancy rates. The combined OR is 1.70 (95% CI 1.30, 2.23, p < 0.05) [Benchaib et al. 2007; Boe-Hansen et al. 2006; Borini et al. 2006; Bungum et al. 2007; Filatov et al. 1999; Frydman et al. 2008; Henkel et al. 2003; Host et al. 2000; Huang et al. 2005; Lin et al. 2008; Tarozzi et al. 2009]. The characteristics of these 11 IVF studies included in the analysis are shown in Tables 2 and 3. Eleven otherwise valid studies must be excluded from meta-analysis because they either include a mixed protocol (IVF and IVF/ICSI; see below) [Larson-Cook et al. 2003; Payne et al. 2005; Seli et al. 2004; Velez de la Calle et al. 2008; Virro et al. 2004] or a 2 × 2 analysis table could not be constructed because a cutoff level for DNA damage was not reported [Bakos et al. 2008; Gu et al. 2009; Meseguer et al. 2008; Morris et al. 2002; Tomlinson et al. 2001; Tomsu et al. 2002]. The results of this updated meta-analysis on IVF studies suggest that sperm DNA and chromatin damage has a modest influence on pregnancy rates at IVF and are in keeping with the results of a more limited meta-analysis [Collins et al. 2008].

Table 2.  Selected Diagnostic Properties of 11 Studies on Sperm DNA Damage and Pregnancy After IVF.

Study	n	Assay	%hDD	Sens	Spec	PPV	NPV	OR	(95% CI)
Filatov et al. 1999	176	CC	41	0.46	0.88	0.96	0.21	6.34	(1.82, 22.08)
Host et al. 2000	175	TUNEL	30	0.34	0.79	0.77	0.37	1.92	(0.92, 4.04)
Henkel et al. 2003	208	TUNEL	69	0.35	0.81	0.81	0.35	2.24	(1.09, 4.58)
Huang et al. 2005	217	TUNEL	19	0.22	0.83	0.50	0.57	1.30	(0.66, 2.56)
Boe-Hansen et al. 2006	139	SCSA	5	0.06	0.97	0.86	0.29	2.43	(0.28, 20.83)
Borini et al. 2006	82	TUNEL	16	0.17	0.89	0.85	0.23	1.66	(0.33, 8.28)
Lin et al. 2008	137	SCSA	16	0.15	0.83	0.45	0.51	0.88	(0.35, 2.19)
Benchaib et al. 2007	84	TUNEL	10	0.07	0.86	0.50	0.32	0.46	(0.11, 2.00)
Bungum et al. 2007	388	SCSA	16	0.17	0.86	0.71	0.34	1.24	(0.69, 2.26)
Frydman et al. 2008	117	TUNEL	44	0.58	0.68	0.64	0.35	2.97	(1.39, 6.32)
Tarozzi et al. 2009	82	CMA3	17	0.22	0.97	0.97	0.28	10.86	(0.62, 191.5)

Abbreviations: %hDD = proportion of samples with high sperm DNA damage; Sens = sensitivity; Spec = specificity; PPV = positive predictive value; NPV = negative predictive value; OR = odds ratio; CC = chromatin compaction; TUNEL = terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling; SCSA = sperm chromatin structure assay; CMA3 = chromomycin A3.

Table 3.  Selected Characteristics of 11 Studies on Sperm DNA Damage and Pregnancy After IVF.

Study	Population	Outcome	Design	Female factors	Cutoff based on
Filatov et al. 1999	Unselected	CP	Unspecified	Unspecified	Large population
Host et al. 2000	Unselected	BP	Prospective	Unspecified	Fertile population
Henkel et al. 2003	Unselected	CP	Prospective	Unspecified	ROC analysis
Huang et al. 2005	Unselected	CP	Retrospective	Unspecified	Unspecified
Boe-Hansen et al. 2006	Unselected	CP	Prospective	FSH<10	Evenson et al. 1999
Borini et al. 2006	Unselected	CP	Unspecified	Unspecified	Benchaib et al. 2007
Lin et al. 2008	Normal semen	CP	Prospective	<40 years, FSH<15	Evenson et al. 1999
Benchaib et al. 2007	Unselected	CP	Prospective	Unspecified	IVF & ICSI results
Bungum et al. 2007	Female factor	BP	Prospective	<40 years, FSH<12	Evenson et al. 1999
Frydman et al. 2008	Unselected	CP	Prospective	<38 years, FSH<10	Median value
Tarozzi et al. 2009	Male factor	CP	Unspecified	Unspecified	ROC analysis

Abbreviations: CP = clinical pregnancy; BP = biochemical pregnancy; ROC = receiver operating characteristics.

In Vitro Fertilization With Intracytoplasmic Sperm Injection (IVF/ICSI) Studies

Over 20 studies have evaluated the relationship between sperm DNA integrity and pregnancy rates after IVF/ICSI. In this review, using the above criteria, only 14 can be evaluated. As with IVF studies, these ICSI studies are quite heterogeneous in terms of design, inclusion/exclusion criteria, and the sperm DNA damage test employed. In keeping with a recent analysis [Collins et al. 2008], the results of this updated meta-analysis on ICSI studies indicate that sperm DNA damage is not related to ICSI pregnancy rates (combined OR of 1.15, 95% 0.90, 1.55, p = 0.65) [Benchaib et al. 2007; Boe-Hansen et al. 2006; Borini et al. 2006; Bungum et al. 2007; Check et al. 2005; Gandini et al. 2004; Hammadeh et al. 1996; Henkel et al. 2003; Host et al. 2000; Huang et al. 2005; Lin et al. 2008; Micinski et al. 2009; Tarozzi et al. 2009; Zini et al. 2005]. The characteristics of the 14 ICSI studies that could be evaluated are shown in Tables 4 and 5. Nine otherwise valid studies were removed from the meta-analysis because: (i) they included a mixed population (IVF and IVF/ICSI) [Larson-Cook et al. 2003; Payne et al. 2005; Seli et al. 2004; Velez de la Calle et al. 2008; Virro et al. 2004], (ii) a 2 × 2 table could not be constructed [Bakos et al. 2008; Morris et al. 2002; Muriel et al. 2006], or (iii) the assay type utilized in the study is not widely recognized [Virant-Klun et al. 2002]. Although it may be surprising that studies indicate that sperm DNA and chromatin damage is not related to ICSI pregnancy, it is suspected that the careful selection of the sperm and embryo during human ICSI abrogates the likely adverse effect(s) of sperm DNA damage on reproductive outcomes [Gandini et al. 2004]. However, we cannot exclude the possibility that very high levels of DNA damage will impact pregnancy outcome. Perhaps the most concerning aspect of these findings is the unknown long-term consequence (i.e., post-natal health) of a successful pregnancy with very high levels of DNA damage.

Table 4.  Selected Diagnostic Properties of 14 Studies on Sperm DNA Damage and Pregnancy After ICSI.

Study	n	Assay	%hDD	Sens	Spec	PPV	NPV	OR	95% CI
Hammadeh et al. 1996	60	ABlue	44	0.50	0.71	0.82	0.35	2.40	(0.72, 7.96)
Host et al. 2000	61	TUNEL	59	0.57	0.38	0.58	0.36	0.79	(0.28, 2.25)
Henkel et al. 2003	54	TUNEL	48	0.68	0.63	0.79	0.50	3.67	(1.12, 12.0)
Gandini et al. 2004	22	SCSA	41	0.31	0.44	0.44	0.31	0.36	(0.06, 2.08)
Huang et al. 2005	86	TUNEL	57	0.64	0.50	0.55	0.60	1.80	(0.76, 4.27)
Zini et al. 2005	60	SCSA	18	0.17	0.81	0.46	0.51	0.87	(0.23, 3.22)
Check et al. 2005	104	SCSA	28	0.29	0.76	0.72	0.34	1.34	(0.52, 3.43)
Boe-Hansen et al. 2006	47	SCSA	38	0.36	0.57	0.67	0.28	0.76	(0.21, 2.72)
Borini et al. 2006	50	TUNEL	60	0.71	0.75	0.90	0.45	7.36	(1.67, 32.4)
Benchaib et al. 2007	218	TUNEL	17	0.19	0.87	0.72	0.37	1.55	(0.70, 3.41)
Bungum et al. 2007	223	SCSA	33	0.29	0.61	0.52	0.37	0.65	(0.37, 1.14)
Lin et al. 2008	86	SCSA	24	0.26	0.77	0.52	0.52	1.21	(0.45, 3.23)
Micinski et al. 2009	50	SCSA	35	0.40	0.85	0.91	0.28	3.73	(0.74, 18.77)
Tarozzi et al. 2009	50	CMA3	56	0.49	0.27	0.61	0.18	0.34	(0.09, 1.29)

Abbreviations: %hDD = proportion of samples with high sperm DNA damage; Sens = sensitivity; Spec = specificity; PPV = positive predictive value; NPV = negative predictive value; OR = odds ratio; ABlue = aniline blue; TUNEL = terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling; SCSA = sperm chromatin structure assay; CMA3 = chromomycin A3.

Table 5.  Selected Characteristics of 14 Studies on Sperm DNA Damage and Pregnancy After ICSI.

Study	Population	Outcome	Design	Female factors	Cutoff based on
Hammadeh et al. 1996	Male factor	CP	Prospective	Unspecified	ROC analysis
Host et al. 2000	Male factor	CP	Prospective	Unspecified	Fertile population
Henkel et al. 2003	Unselected	CP	Prospective	Unspecified	Large population
Gandini et al. 2004	Unselected	CP	Prospective	Unspecified	Evenson et al. 1999
Huang et al. 2005	Unselected	CP	Retrospective	Unspecified	Unspecified
Zini et al. 2005	Unselected	CP	Prospective	Unspecified	Evenson et al. 1999
Check et al. 2005	IVF failures	CP	Unspecified	Unspecified	Evenson et al. 1999
Boe-Hansen et al. 2006	Unselected	CP	Prospective	Unspecified	Large population
Borini et al. 2006	Unselected	CP	Unspecified	Unspecified	Benchaib et al. 2007
Benchaib et al. 2007	Unselected	CP	Prospective	Unspecified	IVF & ICSI results
Bungum et al. 2007	Male factor	CP	Prospective	<40 years, FSH<12	Evenson et al. 1999
Lin et al. 2008	Male factor	CP	Prospective	<40 years, FSH<15	Evenson et al. 1999
Micinski et al. 2009	Unselected	CP	Prospective	<38 years	Filatov et al. 1999
Tarozzi et al. 2009	Male factor	CP	Unspecified	Unspecified	ROC analysis

Abbreviations: CP = clinical pregnancy; ROC = receiver operating characteristics.

Mixed or Combined IVF and IVF/ICSI Studies

Seven mixed IVF and IVF-ICSI studies (5 evaluable) have assessed the relationship between sperm DNA integrity and pregnancy. Two otherwise valid studies were removed from the meta-analysis because a 2 × 2 table could not be constructed [Tavalaee et al. 2009; Velez de la Calle et al. 2008]. Again, these combined studies are quite heterogeneous in terms of design, inclusion/exclusion criteria, and the sperm DNA damage test employed. A systematic review and meta-analysis of the 5 mixed IVF and IVF-ICSI studies shows that in this context sperm DNA damage correlates with the rate of pregnancy (combined OR of 1.63, 95% 1.03, 2.59, p < 0.05) [Larson-Cook et al. 2003; Meseguer et al. 2008; Payne et al. 2005; Seli et al. 2004; Virro et al. 2004]. The results of this meta-analysis on mixed IVF and IVF-ICSI studies are in keeping with the above meta-analysis of IVF studies and support the premise that sperm DNA damage has a measurable but modest impact on pregnancy rates at IVF.

Pregnancy Loss After IVF and IVF-ICSI

Eleven studies have described sperm DNA damage and its association with pregnancy loss after IVF and IVF/ICSI (see Table 4). These pregnancy loss studies are quite heterogeneous in terms of design, inclusion/exclusion criteria, and the sperm DNA damage test employed. A meta-analysis of the studies that could be examined demonstrates a combined OR of 2.48 (95% CI; 1.52, 4.04, p < 0.0001) indicating an important association between sperm DNA damage and the rate of pregnancy loss after IVF and ICSI [Zini et al. 2008]. The characteristics of these studies, from 7 reports, are shown in Table 6. There was no difference in the OR according to the type of ART (IVF or ICSI). These data provide an additional mechanism or cause for pregnancy loss after IVF and IVF/ICSI. It may be speculated that the pregnancy loss (at IVF or ICSI) is a result of impaired embryo/blastocyst development associated with sperm DNA damage [Seli et al. 2004; Zini et al. 2005].

Table 6.  Selected Diagnostic Properties of Studies on Sperm DNA Damage and Pregnancy Loss (PL) After IVF and IVF/ICSI.

Study	n	ART	Assay	PL (%)	Ab Test* (%)	Sens	Spec	PPV	NPV	OR (95% CI)
Check et al. 2005	104	ICSI	SCSA	47	24	0.31	0.83	0.63	0.58	2.27 (0.45, 11.59)
Zini et al. 2005	60	ICSI	SCSA	16	19	0.40	0.85	0.33	0.88	3.67 (0.46, 29.42)
Borini et al. 2006	82	IVF	TUNEL	6	11	0.91	0.94	0.50	0.99	160 (0.18,144708)
Borini et al. 2006	50	ICSI	TUNEL	25	25	0.97	0.99	0.97	0.99	2700 (0.38, 2×10^7)
Benchaib et al. 2007	84	IVF	TUNEL	15	15	0.50	0.91	0.50	0.91	10.0 (0.87, 114.8)
Benchaib et al. 2007	218	ICSI	TUNEL	12	15	0.38	0.88	0.30	0.91	4.54 (0.89, 23.28)
Bungum et al. 2007	388	IVF	SCSA	24	14	0.11	0.85	0.19	0.76	0.73 (0.23, 2.33)
Bungum et al. 2007	223	ICSI	SCSA	19	40	0.50	0.63	0.24	0.84	1.69 (0.63, 4.49)
Frydman et al. 2008	117	IVF	TUNEL	19	32	0.64	0.75	0.37	0.90	5.25 (1.31, 21.11)
Lin et al. 2008	137	IVF	SCSA	10	17	0.29	0.84	0.17	0.92	2.16 (0.37, 12.72)
Lin et al. 2008	86	ICSI	SCSA	18	23	0.50	0.83	0.40	0.88	5.00 (0.97, 25.77)

Abbreviations: ART = assisted reproductive technology; PL = pregnancy loss; Ab Test = proportion of abnormal sperm DNA test amongst documented pregnancies; Sens = sensitivity; Spec = specificity; PPV = positive predictive value; NPV = negative predictive value; OR = odds ratio.

Clinical value of tests of sperm DNA and chromatin damage

Four common clinical scenarios are highlighted that demonstrate the clinical value of sperm DNA damage and chromatin integrity tests. The recommendations for sperm DNA testing are based on: (i) a systematic review and meta-analysis of the relevant studies, (ii) the characteristics of sperm DNA testing (e.g., sensitivity, positivity rate), and (iii) on disease prevalence (e.g., pregnancy, pregnancy loss).

1. Screening Test for First Pregnancy Planners

The data from 3 studies show that sperm DNA damage is associated with a significantly reduced natural pregnancy rate (combined OR 7.01, 95% CI 3.68, 13.36, p < 0.0001). Remarkably, the 3 studies reported very similar associations between sperm DNA damage and natural pregnancy rate (with ORs of 6.54, 6.82, and 7.59). An analysis of the 3 studies reveals a median pregnancy rate of 53%, with a median positive predictive value (PPV) of 83% and a median negative predictive value (NPV) of 58% associated with sperm DNA testing [Evenson et al. 1999; Giwercman et al. 2010; Spano et al. 2000]. As such, the analysis predicts that in populations presenting an overall pregnancy rate of 53% (at 6 to 12 months of follow-up), the pregnancy rate is 17% when there is a positive test for sperm DNA damage and 58% when the test result is normal. Therefore, sperm DNA damage testing can discriminate between pregnancy rates of 17% and 58%. However, because the prevalence of a positive test in this context (first pregnancy planners) is low (<10%) and 17% of couples with a positive test will achieve a pregnancy, indiscriminate sperm DNA testing in this context is not advocated. Clinicians may want to test first pregnancy planners but clinicians should also understand the predictive value and limitations (e.g., sensitivity, specificity) of the sperm DNA test in this context and discuss these issues with the patients.

2. Couples With Mild Male-factor Infertility IUI Candidates

Data from one valid IUI study shows that sperm DNA damage is related to a significantly reduced IUI pregnancy rate (OR 9.9, 95% CI, 2.37, 41.51, p < 0.0001). In the Bungum et al. [2007] study, the overall IUI pregnancy rate is 20%, the PPV is 97%, and the NPV is 24%. Therefore, in populations with an IUI pregnancy rate of 20%, a positive test for sperm DNA damage predicts the pregnancy rate to be 3% and a normal test result predicts the pregnancy rate to be 24%. Therefore, assessing sperm DNA damage prior to IUI can differentiate between pregnancy rates of 3% and 24%. According to the Bungum et al. [2007] study, couples with high levels of sperm DNA damage should proceed to IVF and/or ICSI rather than IUI. However, it is important to note that the sensitivity and prevalence of a positive test in this context (couples with mild male-factor infertility) are low (<20%) and these recommendations are derived from only one reliable study [Bungum, et al. 2007]. As such, additional IUI studies are needed before routine testing is recommended prior to initiating IUI treatments.

3. Couples With Severe Male-factor Infertility IVF or ICSI Candidates

Data from more than 20 studies of which 11 could be assessed demonstrate that sperm DNA damage is associated with a modest but significant reduction in the IVF pregnancy rate (combined OR of 1.70, 95% CI 1.30, 2.23, p < 0.05). Further analysis of the 11 IVF studies (with a median pregnancy rate of 33%) reveals a median PPV of 77% and median NPV of 34% (see Table 3). In populations with an overall IVF pregnancy rate of 33%, a positive test for sperm DNA damage predicts the IVF pregnancy rate of 23% and if the test is negative, 34%. As such, couples with sperm DNA damage may choose to proceed to ICSI, where pregnancy rates are independent of the test results (combined OR of 1.15, 95% 0.90, 1.55, p=0.65; see Table 4). The clinical value of an 11% difference in pregnancy rates (34% vs. 23%) is modest and it may be hard to justify routine testing. However, clinicians may want to test select couples (e.g., with failed IVF) so as to better counsel these couples in future ART cycles.

Testing couples with severe male-factor infertility may also be valuable because sperm DNA damage is associated with a significantly higher rate of pregnancy loss after IVF or ICSI (combined OR of 2.48, 95% CI; 1.52, 4.04, p < 0.0001). Data derived from these studies (PPV and NPV) indicate that in populations exhibiting an 18% rate of pregnancy loss, the rate of pregnancy loss is estimated at 37% when the test is positive and 10% when it is negative. The difference between a pregnancy loss rate of 37% and 10% may be valuable to patients and clinicians. Although the effect of DNA damage on pregnancy loss should be discussed with patients prior to undergoing ART, many couples will proceed with these treatments regardless of sperm DNA test results and the impact on pregnancy loss.

4. Couples With Pregnancy Loss After IVF or IVF/ICSI

The prevalence of a positive test, sensitivity, and specificity of sperm DNA testing in the context of pregnancy loss after IVF and ICSI are 25%, 40%, and 85%, respectively. This indicates that sperm DNA damage is a minor cause of pregnancy loss after IVF and ICSI (based on the low prevalence and low sensitivity). However, if the test is positive, it suggests that the sperm DNA damage (or male-factor) may be the cause of the pregnancy loss (based on the high specificity). In this setting it may be advisable to evaluate or re-evaluate the male and correct any potential male factor, like varicocele, that may contribute to the DNA damage.

Summary

Tests of sperm DNA and chromatin integrity are being used in the evaluation of the infertile man with the hypothesis that these tests may better diagnose the infertility and predict reproductive outcomes. To date, the clinical studies on sperm DNA and chromatin defects allow us to conclude that sperm DNA damage is associated with lower natural, IUI, and IVF pregnancy rates, but not with ICSI pregnancy rates. However, it is associated with an increased risk of pregnancy loss. Although the clinical utility of these tests for sperm DNA/chromatin damage remains to be firmly established, the data suggest that there is clinical value in testing couples with recurrent miscarriages, or prior to initiating ART cycles.

Declaration of Interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Abbreviations

IVF:	=	in vitro fertilization
ICSI:	=	intra-cytoplasmic sperm injection
IUI:	=	intra-uterine insemination
ARTs:	=	assisted reproductive technologies
ROS:	=	reactive oxygen species
OR:	=	odds ratio
PPV:	=	positive predictive value
NPV:	=	negative predictive value.