Prevalence of high DNA fragmentation index in male partners of unexplained infertile couples



Krzysztof Oleszczuk, Reproductive Medicine Centre, Skåne University Hospital, Jan Waldenströms gata 47, Malmö 205 02, Sweden. E-mail:


The sperm chromatin structure assay (SCSA) parameter DNA fragmentation Index (DFI) is a valuable tool for prediction of fertility in vivo. Clinical data show that a DFI above 30% is associated with very low chance for achieving pregnancy by natural conception or by insemination. Already when DFI is above 20% the chance of natural pregnancy is reduced, this despite normal conventional semen parameters. The aim of the present study was to investigate the prevalence of high DFI in male partners of unexplained infertile couples to further identification of male factors contributing to subfertility. Among 212 consecutive men under infertility investigation, 122 cases with the diagnosis ‘unexplained infertility’ were identified. For all but three, SCSA data were available. The percentage of couples with diagnosis ‘unexplained infertility’ in which the male partner has DFI >20% or DFI >30% was calculated. In the group diagnosed with ‘unexplained infertility’ 17.7% of the men (95% CI 10.8–24.5) presented with 20 ≤DFI <30 and 8.4% (95% CI 3.40–13.4) had DFI ≥30%. A significant part of men diagnosed as unexplained infertile according to traditional diagnostic methods has remarkably high degrees of fragmented sperm DNA. Apart from adding to our understanding of biology of infertility our finding has clinical implications. Couples in which the DFI of the male partner is high can avoid prolonged attempts to become spontaneously pregnant or referral for intrauterine insemination, both having low chances of leading to conception.


Infertility is a common problem that affects up to 25% of couples in societies in various parts of the world (Schmidt et al. 1995; Bushnik et al. 2012; Cai et al. 2011; Dunson et al. 2004). The exact prevalence of male factor infertility is difficult to define referable to the lack of sufficient diagnostic tools Jequier 2004). Although the World Health Organization (WHO, 1987) has estimated that up to 50% of the infertility cases are predominantly or partly caused by male factors, the incidence of infertile couples diagnosed as unexplained infertile is around 10–20% (Isaksson & Tiitinen 2004).

Investigation of the male partner in infertile couple is mainly based on the conventional semen analysis, which includes assessment of sperm concentration, motility and morphology. These parameters have, however, a limited power in regard to prediction of chance of conception (Bonde et al. 1998) and can only in selected cases point to options for specific therapeutic measures. To overcome these limitations, a number of new sperm tests have been developed (Erenpreiss et al. 2006). The sperm chromatin structure assay (SCSA), first described by Evenson (Evenson et al. 1980) evaluates sperm chromatin integrity and provides additional information about the fertilizing capacity of the sperm. Studies have shown that the SCSA parameter DNA fragmentation index (DFI) is an independent predictor of male sub-fertility in vivo (Bungum et al. 2007), Giwercman et al. (2010). Recently we demonstrated that men having normal standard semen parameters and an increased DFI above 20% had a higher odds ratio for infertility compared with fertile controls (Giwercman et al. 2010). If one of the standard semen parameters according to World Health Organization criteria was abnormal (WHO 1999), the odds ratio for infertility increased already at DFI above 10%. Thus, chances of conception achieved by intercourse or by intra-uterine insemination decreased already at DFI levels above 20% and are being close to zero when DFI exceeds the level of 30% (Giwercman et al. 2010). These findings indicate that DFI is a potentially, clinically useful marker of male fertility as it can add to explaining, at least some cases of ‘unexplained infertility’. Clinically, DFI can be of help in selecting couples who, referable to low in vivo fertility potential, should be referred directly for in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI). Furthermore, it has been suggested that high DFI is a potentially curable condition and causal treatment may become an option for cases of infertility associated with impairment of sperm DNA integrity (Agarwal et al. 2009; Li et al. 2012).

So far, there is only limited information regarding the prevalence of high DFI in couples diagnosed with ‘unexplained infertility’. The purpose of the study was, therefore, to find out the percentage of couples with diagnosis ‘unexplained infertility’ in which the male partner has a DFI >20% or a DFI >30%. Furthermore, we wished to compare this proportion with the corresponding figure in a cohort of proven fertile men with normal standard sperm parameters (Giwercman et al., 2010).

Materials and methods

Study design and patient population

This is a case series study based on data from files of 212 consecutive couples who underwent infertility investigation at the Reproductive Medicine Centre (RMC), Skåne University Hospital, Malmö, Sweden between June 2008 and April 2011. Reproductive Medicine Centre is a tertiary referral centre; however, the couples can refer themselves after more than 1 year of unprotected intercourse not leading to pregnancy. As cases with obvious male or female factor are usually referred directly to RMCs andrological or gynaecological outpatient clinic from secondary referral level, couples with ‘unexplained infertility’ are over-represented in this group.

The diagnosis of ‘unexplained infertility’ was based on the following

  • At least 1 year of unprotected intercourse without pregnancy;
  • Normal sperm concentration, motility and morphology according to WHO, 1999;
  • Unremarkable andrological history (no cryptorchidism, drug abuse, cancer treatment or other iatrogenic factors), no genetic abnormalities such as Klinefelter's syndrome or Y-chromosome microdeletion and no hypogonadotropic hypogonadism;
  • No female factors (anovulation, hormonal infertility, tubal factor or endometriosis).

Among the 212 couples included, 27 couples had a female related infertility diagnosis (anovulation, hormonal infertility, tubal factor or endometriosis) and were excluded from the study. The same was true for additional 63 couples with ‘male factor infertility’, defined as one or more abnormal standard sperm parameters. All, except three men, who only had one ejaculate investigated, delivered at least two semen samples for analysis according to WHO criteria (WHO 1999). The SCSA analysis is a routine test for all male patients in our clinic. However, among the 122 ‘unexplained infertile’ couples only 119 (97%) had a SCSA analysis and could thus be included in the data analysis.

For comparison, retrieving data from a previous publication (Giwercman et al., 2010) we included a cohort of 95 proven fertile men with normal standard sperm parameters. Among 95 of these men, 10 presented with DFI ≥20%.

Semen samples and standard semen analysis

Semen samples were collected by masturbation after the recommended abstinence period of 2–7 days. Semen parameters were scored according to the WHO guidelines (WHO 1999). For assigning semen quality as normal, following cut-off levels, which were valid at the time of the collection of our material, were used:

  • Volume ≥2.0 mL;
  • Sperm concentration ≥20 × 106/mL or total number ≥40 × 106;
  • Sperm motility: ≥25% rapidly progressive motile or ≥50% progressively motile sperm;
  • Sperm morphology: ≥5% normal forms.

Sperm chromatin structure assay

The principles and procedure to measure sperm DNA damage using flow cytometry SCSA are described in detail elsewhere (Bungum et al. 2004; Evenson & Jost 2000; Spano et al. 2000). In brief, the SCSA is based on the phenomenon that a 30 sec treatment with pH 1.2 buffers denatures the DNA at the sites of single- or double-strand breaks, whereas normal double-stranded DNA remains intact. Thereafter, the sperm cells are stained with the fluorescent DNA dye acridine orange, which differentially stains double- and single-stranded DNA. After blue light excitation in a flow cytometer, the intact (double-stranded) DNA emits green fluorescence, whereas denatured (single-stranded) DNA emits red fluorescence. Sperm chromatin damage is quantified using the flow cytometry measurements of the metachromatic shift from green (native, double-stranded DNA) to red (denatured, single-stranded DNA) fluorescence and displayed as red vs. green fluorescence intensity cytogram patterns. The extent of DNA denaturation is expressed as DFI, which is the ratio of red to total fluorescence intensity that is, the level of denatured DNA over the total DNA.

A total of 5–10 000 cells were analysed by FACSort (Becton Dickinson, San Jose, CA, USA). Analysis of the flow cytometric data was carried out using dedicated software (SCSASoft; SCSA Diagnostics, Brookings, SD, USA), which imply that the DFI histogram is used to precisely determine the percentage of DFI. All SCSA measurements were performed on raw semen, which on the day of analysis was quickly thawed and analysed immediately. For the flow cytometer setup and calibration, a reference sample was used from a normal donor ejaculate retrieved from the laboratory repository (Evenson & Jost 2000). The same reference sample was used for the whole study period. A reference was run for every fifth sample. The intra-laboratory CV for DFI analysis was found to be 4.5%. A single SCSA measurement was made for each reference sample.

Statistical analysis

Results were expressed as percentage of men with 20 ≤ DFI < 30% and DFI ≥30%, respectively, in relation to the total number of couples diagnosed with ‘unexplained infertility’. The rationale for using these DFI thresholds was based on previous reports in which the SCSA was performed (Bungum et al. 2007; Evenson & Jost 2000; Giwercman et al. 2010). A 95% confidence interval (CI) was estimated for each group. The data analysis was performed on the first semen analysis in which SCSA was performed.

The additional parameters: age of man/woman and woman's body mass index (BMI) were expressed as mean/median (range) separately for each group. These values were also calculated for conventional semen parameters (sperm concentration, motility A + B) for those 119 semen samples with DFI included in the analysis.

Using Fisher's exact test (, the percentages of men with DFI ≥20%, was compared to a corresponding figure in the previously reported cohort of proven fertile men with normal standard sperm parameters (Giwercman et al. 2010). All other statistical analyses were performed using Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA, USA).


In Table 1 the demographic characteristics of the 119 included couples with ‘unexplained infertility’ are given.

Table 1. Background characteristics of the 119 couples with ‘unexplained infertility’. Semen parameters based on the first sample delivered by the patient
 ‘unexplained infertility’
  1. Motility A – rapid progressive motility.

  2. Motility B – slow progressive motility.

Age man (years), mean/median (range)34/33 (22;55)
Age woman (years), mean/median (range)31/31.5 (21;39)
BMI woman (kg/m2), mean/median (range)25/24 (18.5;43.4)
Sperm concentration (×106/mL), mean/median (range)93/73 (15;640)
Sperm motility (A + B) (%), mean/median (range)60/61 (25;81)
DFI (%), mean/median (range)16/15 (4.0;50)

The mean DFI was 16.2% (median 15%, range 4–50%). Twenty one of these men (17.7%) (95% CI 10.8–24.5%) presented with 20 ≤ DFI < 30% and 10 men [8.4%, (95% CI 3.40–13.4%)] had a DFI ≥30%. In total, 31 men [26.1%, (95%CI 18.2–33.9%)] had a DFI ≥20%.

The percentage of men with DFI ≥20%, in the cohort of fertile men with normal standard sperm parameters was 10.5% (95% CI 6.29–17.0%), this value being significantly lower than those found in men from ‘unexplained infertility couples’ (p = 0.005).


The present study shows that one quarter of men in couples diagnosed as ‘unexplained infertile’ according to traditional diagnostic methods have a DFI level ≥20%, previously found to be associated with a decreased fertility in vivo. This figure was statistically significantly higher than in proven fertile men. In a retrospective study (Giwercman et al. 2010) found that 10.5% of men with proven fertility had a DFI level of 20% or higher. Thus, in a significant proportion of so called ‘unexplained’ cases impairment of sperm DNA integrity can at least partly explain the subfertility problem of the couple. In line with previous accumulated data (Bungum et al. 2011) our results suggest that sperm DNA integrity assessment may help to differentiate men with fertility problems and can therefore be of help in counselling of infertile couples.

Recent research has indicated that sperm chromatin integrity testing as assessed with SCSA may contribute to the evaluation of men in infertile couples, however, none of these previous studies have been specifically related to the diagnosis ‘unexplained infertility’ (Giwercman et al. 2010). Previously we reported that if sperm concentration, motility and morphology were normal, fertility impairment is seen at DFI levels exceeding 20% (Spano et al. 2000). It has also been shown in studies based on pregnancy planners (Spano et al. 2000) and on couples referred for intrauterine insemination (Bungum et al. 2007) that the probability of conception in vivo decreases when the DFI, as determined by SCSA, exceeds 20% and is almost zero if this value is more than 30%. This was the reason for selecting ‘cut off’ values of 20% and 30% respectively. Numerous studies have demonstrated that the association between SCSA and other semen parameters is only weak to moderate (Giwercman et al. 2003; Spano et al. 1998). This indicates that impairment of sperm DNA integrity is an independent predictor of male fertility (Bungum et al. 2007; Giwercman et al. 2010).

Sperm DNA integrity assessment has been suggested as being useful in the clinical guidance in choice of assisted reproduction technique (Boe-Hansen et al. 2006; Bungum et al. 2004; Jiang et al. 2011; Zini et al. 2001), although some disagreement regarding this matter exists (Lin et al. 2008). Data indicate that in cases with DFI above 30% the ‘baby take home rate’ is higher when using ICSI instead of standard IVF (Bungum et al. 2011).

One limitation of this study is the possibility to exclude female sub-fertility as a factor contributing to the infertility of the couple. Today, the work up of the female partner in an infertile couple is rather sparse (Crosignani & Rubin 2000), often limited to hormonal evaluation only. Even though we have excluded female factors such as endometriosis, tubal occlusion or ovulatory disturbances, other causes of female subfertility, as for example poor oocyte quality cannot be excluded. However, as infertility, in many cases, is believed to be ascribable to accumulation of several adverse factors, even in case of presence of some ‘female factor’, the contribution of impairment of sperm DNA integrity may play an important role.

The calculations are based on one SCSA analysis only. However, despite some intra-individual variation in the DFI, we have shown (Oleszczuk et al. 2011) that in 85% of cases when repeating SCSA- analysis the DFI value remained at the same side of the 30% cut-off level. Thus, multiple SCSA testing only rarely impels a change of DFI category from normal to abnormal, or vice versa.

Our study has biological and clinical implications. From a biological point of view, it is interesting that sperm DNA impairment can, at least partly, explain as many as 25% of previously unexplained cases. Clinically, our data indicate that SCSA testing may help in management of couples with unexplained infertility. It has been suggested that some cases of impairment of sperm DNA are potentially curable (Agarwal et al. 2009; Li et al. 2012). Furthermore, finding of high DFI will in incurable cases point to direct referral to IVF or ICSI, instead of continuing attempts to achieve spontaneous pregnancy or using intrauterine insemination.

Authors' contributions

K. O, A. G and M. B have all given substantial contributions to conception and design of the present study. All authors have contributed to acquisition of data, analysis as well as interpretation of data. K. O has drafted the manuscript and A. G and M. B have revised the content critically. All authors have made final approval of the version to be published.

Funding statements

The study was economically supported by grants from Swedish Research Council and the Governmental Funding for Clinical Research (ALF) as well as Skåne University Hospital Funds.

Conflict of interests

The authors declare that they have no competing interests.