Sperm DNA fragmentation and its impact on pregnancy outcomes: a narrative review focusing on threshold definition and clinical strategies Authors: Sinhang Cheng, Yuan Yuan, Zihan Wang, Zhuoheng Xie, Qianxi Chen, Zhen Liu, Yan Chen, Tiangang Song, Mengqing Yan, Zhongjie Zheng, Haocheng Lin Published: 2026 ABSTRACT AbstractBackground and ObjectiveSperm DNA fragmentation (SDF) is a key functional indicator for evaluating male fertility and the integrity of genetic material. However, a unified intervention threshold for SDF remains lacking, primarily due to variations in detection methodologies and significant heterogeneity across studies. This narrative review aims to summarize the association between SDF and pregnancy outcomes, with a focus on threshold definition and the influence of different detection methods and clinical scenarios.MethodsA comprehensive literature search was conducted in PubMed and Scopus databases from January 1990 to December 2023. Search terms included “sperm DNA fragmentation”, “DFI”, “SDF”, “male infertility”, “assisted reproductive technology”, “pregnancy outcome”, and “recurrent pregnancy loss”. Studies evaluating the association between SDF and reproductive outcomes, as well as those reporting on diagnostic or prognostic thresholds, were considered for inclusion.Key Content and FindingsHigh SDF is significantly associated with lower fertilization rates and fewer high-quality embryos in assisted reproductive technology (ART), as well as an increased risk of recurrent pregnancy loss (RPL). Its clinical impact varies by mode of conception: it negatively affects natural pregnancy and intrauterine insemination (IUI), whereas in in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI), its effect on fertilization rate is less pronounced, but it may impair subsequent embryo development, reducing blastocyst formation rate and increasing miscarriage rate. The use of testicular sperm for ICSI in men with elevated SDF has been reported in observational studies, with some showing improved pregnancy and live birth rates. A 30% DNA fragmentation index (DFI) [sperm chromatin structure assay (SCSA)/terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL)] is the most frequently referenced threshold, but reported cut offs vary from 10% to 36% depending on assay and outcome. Regarding detection technology, the SCSA is frequently described as the most extensively validated method, while the integration of novel technologies such as artificial intelligence offers the potential for more precise and efficient assessment in the future.ConclusionsSDF should be considered an important supplement to conventional semen analysis, especially in cases of unexplained infertility, ART failure, and RPL. This review emphasizes the necessity of establishing a unified clinical threshold, and current evidence highlights the need to standardize testing protocols, establish evidence-based SDF thresholds, and develop individualized management strategies.Keywords: Sperm DNA fragmentation (SDF), DNA fragmentation index, clinical threshold, male infertility, assisted reproductive technologies Introduction Background Global fertility rates have been declining, from 4.9 in the 1950s to 2.3 in 2023, and further to 2.2 in 2025 (1). The fertility rate required to maintain a stable population is approximately 2.1, and most countries are currently below this level, indicating profound demographic shifts. In this context, the clinical importance of infertility is increasingly prominent. According to the latest definition by the World Health Organization (WHO), infertility is a disease characterized by the failure to establish a clinical pregnancy after 12 months of regular, unprotected sexual intercourse (2). Statistics show that approximately one in seven couples faces difficulty conceiving (3). While infertility has many potential causes, including hormonal imbalances and irregular ovulation in both men and women, male factors account for about 40% of cases (4). However, up to 25% of male partners are not adequately evaluated in clinical practice, change to this phenomenon is an urgent public health imperative (5-7). Traditional male infertility diagnosis relies on WHO-standardized conventional semen analysis, including parameters such as sperm concentration (≥20×106/mL), motility (progressive motility ≥50%), and morphology (normal forms ≥30%) (8). These parameters form the indispensable cornerstone of initial clinical diagnosis and are of paramount importance for fertility assessment (9). However, this evaluation system has fundamental limitations: it does not directly reflect the sperm’s crucial fertilizing capacity or the integrity of the genetic material, making its efficacy in predicting actual fertility outcomes very limited. Clinically, it is common for couples with normal semen parameters to fail to conceive naturally or experience repeated Assisted Reproductive Technology (ART) failures, suggesting functional defects that routine tests cannot detect. Therefore, assessing sperm DNA integrity is essential for the birth of healthy offspring (10). Clinically, it is recommended to interpret multiple semen parameters in combination to guide treatment. Based on the extent to which sperm quality falls below thresholds, couples may be advised to undergo intrauterine insemination (IUI), in vitro fertilization (IVF), or use donor sperm for the treatment of male factor infertility (11). Sperm DNA fragmentation (SDF), as a marker of chromatin damage, plays an independent and significant role in male infertility and reproductive success (12-19). Thus, SDF is a core indicator for assessing male fertility (20). Elevated SDF can affect fertility through several pathways: firstly, oxidative stress can impair the fertilizing potential of sperm function (21-24), including motility, zona pellucida recognition, acrosome exocytosis, and gamete fusion (25). Secondly, sperm with damaged DNA may retain fertilizing capacity (26), but can transmit damaged genetic material to the embryo. Ultimately, the impact of SDF on reproductive success depends on the dynamic balance between the degree of DNA damage and the oocyte’s DNA repair capacity. When sperm DNA damage exceeds the oocyte’s repair capacity, it can lead to adverse reproductive outcomes, affecting embryo developmental potential and offspring health (27,28). Furthermore, multiple factors are associated with elevated SDF (29,30), including varicocele, chronic diseases, accessory gland infections, advanced paternal age, unhealthy lifestyles, obesity, occupational environmental exposure, medications, ionizing/non-ionizing radiation, and heat exposure, these relations further highlight the clear link between SDF and male infertility. Rationale and knowledge gap Major clinical guidelines hold divergent views on SDF. The 2024 American Urological Association (AUA) guidelines (31) acknowledge that SDF is negatively correlated with pregnancy rates and positively correlated with miscarriage (32), but emphasize that this association is not fully defined due to variations in threshold definitions and testing methods across studies. The 2022 European Society of Human Reproduction and Embryology (ESHRE) consensus (33) recommends SDF testing for diagnostic purposes. The WHO notes that although meta-analyses confirm the impact of SDF on pregnancy outcomes, a unified clinical threshold has not been established due to significant differences in testing methods. Common methods used clinically, such as TUNEL and SCSA, yield thresholds for diagnosing and predicting ART outcomes that range from 15% to 30%, lacking a universally accepted standard (34,35). This makes defining a pathological threshold critical for guiding interventions. Objectives Based on this background, the core objectives of this review are: first, to narratively summarize the published SDF thresholds and their reported associations with ART outcomes. Second, to describe the influence of testing methods and population characteristics on reported threshold values, based on the existing literature. Third, to outline the ranges of SDF levels that have been explored in relation to different clinical strategies (IUI, IVF, ICSI, testicular sperm) as reported in published studies. Through systematic evidence synthesis, this review aims to provide clinicians with decision-making references for treatment strategy selection, prognosis assessment, and advanced interventions, facilitating the translation of SDF from a laboratory indicator to a key parameter that effectively guides clinical pathways. We present this article in accordance with the Narrative Review reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0544/rc). Methods This narrative review was conducted to synthesize the available evidence on the association between SDF and pregnancy outcomes, with a particular focus on threshold definition and clinical applicability. A systematic literature search was performed in the PubMed and Scopus databases. The search strategy is summarized in Table 1. Table 1The search strategy summaryItemsSpecificationDate of searchOctober 15, 2025Databases and other sources searchedPubMed, Scopus, the reference lists of included reviews and guidelines were also screened for additional relevant studiesSearch terms used(“sperm DNA fragmentation” OR “DFI” OR “SDF” OR “sperm chromatin structure assay” OR “SCSA” OR “TUNEL” OR “sperm chromatin dispersion”) AND (“male infertility” OR “assisted reproductive technology” OR “ART” OR “IVF” OR “ICSI” OR “IUI”) AND (“pregnancy outcome” OR “fertilization rate” OR “blastocyst” OR “miscarriage” OR “recurrent pregnancy loss”)—with filters: English language, human subjectsTimeframeJanuary 1990 to December 2023Inclusion and exclusion criteriaIncluded: original research, systematic reviews, meta-analyses, and clinical guidelines published in English, involving human subjects, and reporting on the relationship between SDF and at least one reproductive outcome (fertilization, embryo quality, pregnancy, miscarriage, live birth) or reporting on a specific SDF threshold. Excluded: case reports, conference abstracts, animal studies, and studies not reporting on SDF or reproductive outcomesSelection processTwo authors (Z.W. and Z.X.) independently screened titles and abstracts. Full texts of potentially eligible studies were then assessed independently by the same two authors. Disagreements were resolved through discussion with a third author (S.C.) The search strategy summary Given the heterogeneity in study designs, populations, SDF assays, and outcome definitions, a narrative synthesis approach was adopted. Data from the included studies were summarized descriptively, focusing on reported thresholds, assay characteristics, and clinical scenarios. Historical development of SDF The research on SDF rate began in the 1990s. Its development process profoundly reflects the paradigm shift of male infertility assessment from morphology to functionality, and perfectly interprets the translational medical path of “from laboratory to clinical”. The following table elaborates on the milestones at each time point and their core significance (Table 2). Table 2Developmental milestones of SDF testingTime pointKey event/milestoneCore content and significance1992WHO Manual, 3rd EditionTheoretical foundation: first noted that conventional semen parameters cannot fully explain infertility causes1999Evenson team’s flow cytometry (SCSA)Technological birth: quantitative detection of SDF (1)Early 2000sBaker et al. multicenter studiesClinical correlation: SDF >30% associated with decreased IVF success and increased miscarriage risk (2)2010EAU Guidelines & WHO 5th Ed.Critical turning point/standardization: EAU first included SDF in guidelines; WHO standardized procedures2020ASRM GuidelinesEstablished decision-making value in RPL and ART2021WHO Manual, 6th EditionQuality enhancement: updated QC standards for SDF testingCurrentField ConsensusGold standard established: SDF is the gold standard for sperm genetic integrityART, assisted reproductive technology; ASRM, American Society for Reproductive Medicine; EAU, European Association of Urology; IVF, in vitro fertilization; QC, quality control; RPL, recurrent pregnancy loss; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; WHO, World Health Organization. Developmental milestones of SDF testing ART, assisted reproductive technology; ASRM, American Society for Reproductive Medicine; EAU, European Association of Urology; IVF, in vitro fertilization; QC, quality control; RPL, recurrent pregnancy loss; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; WHO, World Health Organization. The evolution of guidelines for SDF detection of clinical status The evolution of clinical guidelines on SDF testing reflects a clear shift from initial caution to broader clinical acceptance. Early recommendations deemed routine testing unwarranted due to limited evidence and treatment options. Over time, guidelines began acknowledging SDF’s prognostic value in assisted reproduction and its role in guiding interventions such as varicocele repair and testicular sperm retrieval. A key milestone came with specialty guidelines on recurrent pregnancy loss (RPL), which explicitly linked SDF to etiology. Subsequent comprehensive recommendations endorsed SDF testing for both diagnostic clarity and treatment selection in conditions like unexplained infertility and RPL. The most recent updates affirm these clinical utilities while noting ongoing challenges such as methodological variability and lack of standardized thresholds. The following table elaborates on the comparative evolution of stances on SDF testing in major clinical guidelines (Table 3). Table 3Comparative evolution of stances on SDF testing in major clinical guidelinesOrganizationGuideline/yearCore recommendationKey changes/featuresASRM2013 (3)Not recommended as a routine infertility test (insufficient data, lack of effective treatment options)Conservative phaseASRM2015 (3)Began acknowledging clinical value: varicocele repair and antioxidant therapy may help reduce SDF; SDF testing can provide reference information for IUI, IVF, and ICSI outcomes; testicular sperm has better DNA qualityInitial recognition: shift from “not recommended” to “can provide reference information”ESHRE2018 (RPL Guideline) (4)Specifically addressed SDF testing: a clear association exists between SDF and RPL, independent of female factors; testing can be considered for couples with RPL to find etiologySpecialty area breakthrough: first explicit recommendation in an RPL guidelineEAU2020 (5)Most proactive stance: explicitly recommends SDF testing for men with RPL and unexplained infertility; recommends varicocele repair; for ART failure, testicular sperm ICSI can be consideredComprehensive recommendation: clarified the dual value in diagnosis and treatmentAUA/ASRM2024 (6)Confirms SDF’s negative correlation with pregnancy rates and positive correlation with miscarriage; points out issues like inconsistent thresholds and non-standardized testing; testicular sperm ICSI shows superior clinical outcomes (pregnancy rate, live birth rate, miscarriage rate) for high SDF patientsEvidence-based update: acknowledges limitations while affirming the value of specific interventionsART, assisted reproductive technology; ASRM, American Society for Reproductive Medicine; AUA, American Urological Association; EAU, European Association of Urology; ESHRE, European Society of Human Reproduction and Embryology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; RPL, recurrent pregnancy loss; SDF, sperm DNA fragmentation. Comparative evolution of stances on SDF testing in major clinical guidelines ART, assisted reproductive technology; ASRM, American Society for Reproductive Medicine; AUA, American Urological Association; EAU, European Association of Urology; ESHRE, European Society of Human Reproduction and Embryology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; RPL, recurrent pregnancy loss; SDF, sperm DNA fragmentation. Other semen parameters reflecting pregnancy outcomes Other conventional semen parameters—specifically total motile sperm count (TMC) and strict morphology—have also been evaluated for their association with pregnancy outcomes, though their predictive value varies by treatment modality. In summary, TMC demonstrates a clear threshold effect in IUI, with very low success rates below 5–10 million and plateauing benefit above that range, whereas its utility in IVF/ICSI is limited because these techniques largely circumvent severe male factor infertility. Morphology assessed by strict criteria shows only a weak independent association with outcomes; its clinical relevance emerges primarily when combined with a low TMC, in which case both pregnancy and live birth rates decline significantly. Given the significant correlation between DFI and abnormal sperm morphology (36), and the well-established association between morphology and pregnancy outcomes in ART (16,37), the observed impact of morphological defects on reproductive success may be at least partially mediated by concurrent DNA fragmentation. The following table elaborates on the predictive value and thresholds of TMC and strict morphology (Table 4). Table 4Association of other semen parameters with pregnancy outcomesParameterApplicable technologyClinical threshold and association with pregnancy outcomesTMCIUI (7)Threshold range 5–10 million. Pregnancy rate remains only 1.5% for TMC <10 million; 10.5% for TMC 10–30 million (8); lower pregnancy and live birth rates for TMC <5 million (9); pregnancy rates plateau after exceeding 10 million (10)When TMC 1.6–5 million, pregnancy rate is very low (3.6%) (11); when TMC <10 million, IUI pregnancy rate is extremely low, and IVF may be more cost-effective. If IUI motility <30%, pregnancy rate <3.6% (12)TMC >200 million: significant IUI pregnancy rate achieved (13) TMC >30 million: pregnancy rate at 12.0% (8): pregnancy rates are higher when controlled ovarian hyperstimulation is combined with IUI cycles (10)Threshold for IUI effectiveness (12): TMC 1 million, concentration 5×106/mL, progressive motility ≥30% (only one-quarter of WHO normal standards) yields 8% pregnancy rateOptimal IUI threshold (12): TMC 8 million, concentration 40×106/mL yields up to 12.9% pregnancy rate (twice the WHO normal standards)IVF/ICSI (14)TMC has limited value in predicting IVF outcomes. ICSI can make pregnancy rates for severe male factor infertility comparable to other causes (14). Therefore, couples with TMC <10 million are advised to proceed directly with IVF/ICSI. High-quality embryo rate significantly lower (18%) for counts <10×106/mL compared to those ≥10×106/mL (37%) (14)Sperm morphology (strict criteria)IVF/IUINormal sperm morphology ranges from 1% to 39% (15)Kruger study confirmed 4% as the minimum observed value for achieving pregnancy, and 14% as the clinical threshold for ideal IVF outcome. Subsequent studies showed a weak association between morphology and pregnancy outcomes (16)Based on strict morphology categories (≤1%, 2–4%, 5–8%, 9–14%, >14%), results suggest low morphology does not reduce IUI success rates unless the number of inseminated motile sperm is low. However, when morphology is low (≤4%) and insemination TMC is below 5 million, pregnancy and live birth rates decrease (9)ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; TMC, total motile sperm count; WHO, World Health Organization. Association of other semen parameters with pregnancy outcomes ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; TMC, total motile sperm count; WHO, World Health Organization. SDF detection methods SDF detection methods are diverse, each with its own characteristics, including the Comet assay (38,39), sperm chromatin dispersion (SCD) (40), terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) (41), and sperm chromatin structure assay (SCSA) (42,43); SCSA® is frequently described as the most extensively validated and standardized method, providing reliable and highly reproducible quantitative results via flow cytometry, but it requires significant capital investment for the flow cytometer and skilled operators (44). In contrast, other methods are often more cumbersome and time-consuming (45). In 2018, the European Academy of Andrology stated that DNA damage can reduce pregnancy rates, thereby compromising ICSI and IVF cycles, and recommended incorporating DNA damage testing, identifying SCSA and the alkaline Comet assay as the optimal methods (46). The TUNEL method is recommended by AUA/ESHRE and is well-standardized. The SCD test requires no expensive equipment but relies on manual interpretation. The Comet assay can distinguish between single- and double-strand breaks but is operationally complex (47). Other methods like acridine orange staining and aniline blue staining can indirectly reflect DNA fragmentation or chromatin status but are more subjective. In recent years, artificial intelligence (AI) technology has been gradually applied to SDF detection, with deep learning models achieving up to 95% accuracy in predicting SDF values. Computer-Assisted Sperm Analysis (CASA) parameters are significantly correlated with SDF, and combining them with SCD testing can enhance diagnostic reliability. Currently, three primary methods are used clinically to assess SDF: TUNEL, SCSA, and SCD. They differ in principles, sample requirements, and clinical thresholds (48). The following table compares major SDF detection methods in different respective (Tables 5,6). Apoptotic cell subpopulation (ACS), low chromatin stability (LCS), and high chromatin stability (HCS) is concluded in the comparison, ACS reflects the mean level of DNA damage across all evaluated sperm, LCS represents the proportion of sperm with relatively low DNA damage and HCS represents the proportion of sperm with high DNA fragmentation. Table 5Comparison of principles, characteristics, and clinical thresholds of major SDF detection methodsDetection methodSample requirementsClinical threshold (male infertility diagnosis)Clinical threshold (IUI outcome)Clinical threshold (IVF/ICSI outcome)RemarksTUNELWhole ejaculate (2.0–2.5×106 sperm) (17)Flow cytometry >17% (18)Fluorescence >12% (post-wash) (19)Fluorescence >36% (20,21)This method essentially detects DNA at protamine annular junctions. Protocol modifications, such as using DTT for DNA decondensation, are often introduced to enhance sensitivity. Additionally, TUNEL can be used for surgical sperm (e.g., testicular sperm) specimens but is limited to fluorescence microscopy platforms (22)SCSA>500,000 sperm/mL (can be concentrated by centrifugation) (23)>20–25% (23)>25% (24,25)SDF >30–35% (24,25); HDS >25% (24,25)Samples typically need to be frozen and sent to specialized labs for analysis. Clinical thresholds are relatively uniformSCD/Halo Test (26)1–3 million sperm/mL (lower possible) (27)16–30% (optimal 20%) (23) >20% (28)>25% (when using donor oocytes to control female factors) (29)Particularly suitable for low-volume, low-concentration samples. When analyzing testicular tissue (TESA/TESE), double fluorescence staining can be used to distinguish somatic cells from sperm (30)Comet (31)>100 µL semen (≥5,000 sperm) (33)ACS >26%; LCS <74%; HCS >4% (33)ACS >26%; LCS <70%; HCS >2% (33)ACS >27%; LCS <68%; HCS >10% (33)By adjusting experimental conditions, this assay can distinguish break types: alkaline Comet assay detects global SDF, while neutral Comet assay specifically detects double-strand breaks. The more advanced double-tail Comet assay allows simultaneous evaluation and distinction of both break types (34). The Comet assay provides three core quantitative parameters for precise assessment: ACS, representing the average damage level across 100 sperm; HCS, indicating the proportion of sperm with more severe DNA damage; and LCS, reflecting the proportion of sperm with low DNA damageACS, apoptotic cell subpopulation; DTT, dithiothreitol; HCS, low chromatin stability; HDS, high DNA stainability; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; LCS, low chromatin stability; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; TESA, testicular sperm aspiration; TESE, testicular sperm extraction; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Comparison of principles, characteristics, and clinical thresholds of major SDF detection methods ACS, apoptotic cell subpopulation; DTT, dithiothreitol; HCS, low chromatin stability; HDS, high DNA stainability; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; LCS, low chromatin stability; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; TESA, testicular sperm aspiration; TESE, testicular sperm extraction; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Table 6Summary of advantages and disadvantages of different SDF testing methods (35,49)SDF detection methodPrincipleDamage type detectedAdvantagesDisadvantagesAB/TB stainingDyes have increased affinity for loosely packed chromatin or phosphate residues in the sperm nucleusIndirectly reflects abnormal chromatin packaging (protamine deficiency)Fast, simple, low costHigh inter-laboratory variability, poor reproducibility; observer variabilityCMA3 stainingCMA3 competitively binds to DNA, indirectly visualizing DNA with protamine deficiencyIndirectly reflects abnormal chromatin packaging (protamine deficiency)Reliable results, strong correlation with other methodsObserver variabilityTUNELDirectly labels free ends of DNA breaks. Uses TdT to label 3’-OH ends, detected via fluorescence microscopy or flow cytometryDetects both single and double-strand breaksHigh sensitivity, reliable, minimal observer variability; can be assessed by fluorescence microscopy and flow cytometry; can be used on low sperm countsRequires expensive equipment and trained personnel; lacks clear cut-off values; requires inter-laboratory standardizationSCSAMeasures sperm DNA susceptibility to acid denaturation (flow cytometry version of AO test). Acridine orange emits green fluorescence when bound to double-stranded DNA and red fluorescence when bound to single-stranded DNA. Flow cytometry captures fluorescence to calculate SDF and HDSDetects both single and double-strand breaks (sensitive to denaturation)Standardized protocol and rapid; can analyze large cell numbers simultaneously; can distinguish immature sperm cells; reliably estimates percentage of sperm with DNA damageRequires expensive equipment (flow cytometer) and highly skilled technicians; no commercial kitSCD/Halo Test (26,50)Exploits the intrinsic relationship between sperm nuclear DNA integrity and its morphological appearance following nuclear protein depletion. Sperm fixed in agarose (51) are treated with acid (52) to induce DNA denaturation at break sites. Lysis with detergent (50) and reducing agent removes membranes and protamines (17); Intact DNA forms a characteristic halo around the residual nuclear matrix (53); fragmented DNA fails to form this halo (54)Detects both single and double-strand breaksCommercial kit available; no expensive equipment needed (55); easy to perform. SCD is well-suited for low-concentration or low-motility surgical sperm specimens (e.g., PESA, MESA, TESA, TESE) (30). Ideal for low-concentration samples processed by techniques like IMSI, PICSI, MACS, or microfluidicsObserver variability; relatively low number of sperm analyzed (56)Comet assay (57)Based on DNA decondensation, complete removal of nuclear proteins, and single-cell electrophoresis in agarose. Under alkaline or neutral pH, DNA unwinds. Under an electric field, fragmented DNA migrates from the sperm head, forming a “comet tail” visible under fluorescence microscopy (58). The fluorescence intensity of its tail relative to that of its head directly reflects the level of SDF (59)Can distinguish single-strand from double-strand breaks (alkaline/neutral conditions)High sensitivity; can simultaneously distinguish single- and double-strand breaks; suitable for very low sperm countsPoor reproducibility; high inter-observer variability; non-standardized protocols and thresholds; analyzes low cell numbers; requires experienced observersAB/TB, aniline blue/toluidine blue; AO, acridine orange; HDS, high DNA stainability; IMSI, intracytoplasmic morphologically selected sperm injection; MACS, magnetic activated cell sorting; MESA, microsurgical epididymal sperm aspiration; PESA, percutaneous epididymal sperm aspiration; PICSI, physiological intracytoplasmic sperm injection; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; TdT, terminal deoxynucleotidyl transferase; TESA, testicular sperm aspiration; TESE, testicular sperm extraction; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Summary of advantages and disadvantages of different SDF testing methods (35,49) AB/TB, aniline blue/toluidine blue; AO, acridine orange; HDS, high DNA stainability; IMSI, intracytoplasmic morphologically selected sperm injection; MACS, magnetic activated cell sorting; MESA, microsurgical epididymal sperm aspiration; PESA, percutaneous epididymal sperm aspiration; PICSI, physiological intracytoplasmic sperm injection; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; TdT, terminal deoxynucleotidyl transferase; TESA, testicular sperm aspiration; TESE, testicular sperm extraction; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Relationship between SDF and pregnancy outcomes and threshold setting Standard semen parameters remain the cornerstone of male infertility evaluation (60), For men with elevated SDF, although using testicular sperm in men with elevated SDF yields comparable fertilization rates, it shows superior clinical outcomes in terms of clinical pregnancy rate, live birth rate, and reduced miscarriage rate (61). However, these parameters often do not reflect the likelihood of success with ART (62). Additionally, labs use techniques like density gradient centrifugation to select functionally superior sperm subpopulations to improve ART success. However, a fundamental limitation of these assessment systems is their inability to directly reflect the sperm’s crucial fertilizing function, thus limiting their effectiveness in predicting actual fertility outcomes (62), while SDF can fill this gap. Unfortunately, due to various factors including inconsistent thresholds for defining normal/abnormal ranges, non-standardized protocols, use of different tests measuring unrelated parameters for SDF assessment, and a lack of randomized controlled trials (31), SDF research findings are not always consistent, which may contribute to conflicting results regarding the impact of DNA fragmentation on pregnancy outcomes. SDF can adversely affect the outcomes of ART treatments (63-65) and attempts at natural conception (32,66), including miscarriage rates (67-70). There is currently debate on whether SDF can predict pregnancy outcomes. In the AUA/American Society for Reproductive Medicine (ASRM) Guideline Part I [2020], because no prospective study has directly assessed the impact of SDF testing on the clinical management of infertile couples (i.e., couples who undergo testing have different fertility results compared to couples who do not undergo testing), it was stated that SDF testing should not be performed routinely in the initial evaluation of infertile men (7). Regarding whether different ART modalities affect outcomes based on varying SDF levels, studies have yielded mixed results. For instance, Zurera-Egea et al. (71) found that SDF had little effect on fertilization, blastocyst formation, and implantation rates but compromised embryo quality. In contrast, Wang et al. (72) reported identical post-FET embryo development and clinical outcomes between low and high SDF groups using ICSI or IVF. Virro et al. (73) showed that high DNA stainability (HDS) did not affect blastocyst rate or pregnancy outcomes, that men with SDF ≥30% had more other male factors influencing pregnancy outcomes, and that WHO thresholds could not predict ongoing pregnancy. Association with pregnancy outcomes IVF/ICSI Fertilization rate The influence of sperm DNA damage on fertilization rate remains controversial. Some studies show a negative correlation between DNA fragmentation and fertilization rate (74-76), suggesting that sperm carrying DNA damage struggle to develop to the pronuclear stage (77). In IVF, the proportion of DNA strand breaks is negatively correlated with fertilization rate. However, in ICSI, where a single motile sperm is selected and injected, most studies find no correlation between SDF and fertilization rate (78). Another study indicate that fertilization rate significantly decreases when the SDF exceeds 10% (79,80). Conversely, other studies have not found an adverse effect of DNA damage on fertilization rate (81). For instance, Li et al. showed comparable ICSI fertilization rates between high and low DNA damage groups (79.83% vs. 78.16%) (82). This suggests that sperm DNA damage primarily affects pregnancy outcomes rather than the fertilization process; i.e., sperm with damaged DNA can still fertilize, but are prone to developmental arrest and pregnancy failure during subsequent embryo development (83). Pregnancy rate Regarding pregnancy rates, there is also considerable controversy. Some studies have not found a significant impact of DNA damage on pregnancy rates (84). Gardner et al. (85) found no differences in implantation and pregnancy rates between groups with 16% and 40% fragmentation rates measured by SCSA. While some studies show high SDF affects fertilization rate, it was not related to pregnancy rate: IVF pregnancy rates in the normal SDF group and ICSI pregnancy rates in the high SDF group were 55.3% and 51.2%, respectively, showing no significant difference (80). Furthermore, another study found no statistically significant difference in SCD test values between successful and unsuccessful pregnancy cycles (77). However, the former study simultaneously found DNA fragmentation was negatively correlated with implantation rate (80). In contrast, other scholars have reported significantly reduced pregnancy rates in patients with high DNA damage levels (81,83,86-88), with a lower blastocyst rate and a higher risk of failing to achieve an ongoing pregnancy (73,89). Henkel and Benchaib et al. (86,90) found high SDF to be a negative factor affecting IVF/ICSI pregnancy outcomes, suggesting that although DNA damage does not block fertilization, it causes developmental arrest during the phase of paternal genome activation. Furthermore, the number of sperm with DNA strand breaks is also crucial for pregnancy rates in couples with unexplained infertility (78). Embryo quality and development A stage-specific effect has been observed. Studies show that SDF >15% directly impairs embryo morphokinetics, cleavage speed, and overall quality (91). No clear correlation was found between SDF levels and the morphology, development speed, or fragmentation rate of embryos on day 3 (cleavage stage) (77). However, the paternal effect becomes evident before blastocyst formation (day 5), leading to developmental arrest (92). Blastocyst quality and development have been inversely associated with SDF levels (77). Generally, cleavage-stage embryos that divide faster have a higher potential for blastocyst development and pregnancy (93-98), whereas high SDF levels can cause cell division delays by activating additional DNA repair pathways (99-101), thereby directly compromising the final quality of the blastocyst (102). Given that these SDF-induced differences in embryo quality start to manifest from day 2, with a decrease in the proportion of high-quality embryos and an increase in low-quality embryos by days 3 and 5 (103), and it only became apparent on the 5th day of cultivation. Consequently, some studies have examined whether day-5 blastocyst transfer is associated with improved outcomes in patients with elevated SDF, although this remains an area of ongoing investigation. Therefore, selecting day-5 blastocyst transfer over day-3 cleavage-stage embryo transfer for patients with high SDF may be more beneficial for selecting embryos with the highest developmental potential, thereby improving IVF/ICSI implantation and pregnancy outcomes. This also reasonably explains why fertilization rates and “high-quality” embryo rates decline with increasing levels of DNA fragmentation (80). Miscarriage rate Studies consistently show an association between high SDF and increased miscarriage rates. Robinson et al.’s meta-analysis found that high SDF led to a 2.2-fold increase in miscarriage rate (70). Lin et al. also found that SCSA parameters were not related to IVF/ICSI fertilization rates, embryo quality, or pregnancy rates but might be associated with spontaneous abortion rates (104). A large retrospective study of 6,330 ART cycles further confirmed that DFI ≥30% was associated with significantly higher miscarriage rates and lower infant birth weight compared to DFI <15% (105). IUI outcomes The application of SDF in predicting IUI outcomes remains somewhat controversial, but most studies support its predictive value. In specific subgroup analyses, SDF levels in washed semen samples and male age were identified as independent predictors of outcome (106). Notably, no successful pregnancies occurred in cycles with SDF >12% (106,107). Bungum et al. (108) established clinical thresholds for SDF and HDS (27% and 10%, respectively) and demonstrated that patients below these thresholds had better pregnancy outcomes (109). Wright et al. showed that among 387 IUI cycles in 637 couples, the chance of conception was as low as 3% for patients with SDF exceeding 30% (110). However, while Muriel et al. confirmed a negative correlation between SDF and sperm motility, they could not establish a statistical association between SDF and successful IUI pregnancy (111). Natural pregnancy outcomes DNA damage may ultimately affect natural pregnancy rates. Studies indicate that male fecundity significantly decreases when SDF exceeds 30% (66). Further research established a clear association between SDF and pregnancy outcomes, indicating that SDF in the 20–30% range reduces the chance of natural conception, while exceeding 30% poses a serious threat to successful pregnancy (15). RPL SDF is particularly important in RPL. RPL is a distinct condition from infertility, defined as two or more failed pregnancies (112). The etiology of RPL is diverse, including genetic causes, female anatomical abnormalities, infection, hematological and immune system disorders in the female partner, endocrine issues, and male factors (112,113). High levels of SDF are positively correlated with miscarriage (67,70,87,114-117). Increased sperm DNA damage is associated with a “late paternal effect” during male gene expression activation and is negatively correlated with motility (118), thus potentially increasing miscarriage risk (119). McQueen et al.’s meta-analysis showed that male partners of women with a history of RPL had significantly higher SDF rates compared to partners of fertile control women. In cases with SDF above 15%, the incidence of RPL was as high as 30% (67). Studies by Robinson (70), and Zhao (116) (examining couples undergoing IVF/ICSI), and Zidi-Jrah (115) and Carlini (114) (examining couples with RPL after natural conception), all showed a positive correlation between RPL and SDF. According to the latest AUA guidelines, DNA fragmentation testing should be considered for couples with unexplained RPL, and various treatments, including TESE with ICSI, antioxidant administration, donor sperm, varicocele repair, and/or frequent ejaculation, could be employed (31). The following is a summary of the thresholds of the cited articles, with most of them using 30% as the dividing line to study pregnancy outcomes. However, some articles, based on better data presentation, use independent thresholds. Meanwhile, there are also articles suggesting that a cut-off value of 20% has higher sensitivity (Table 7). Table 7Summary of studies evaluating the association between SDF thresholds and reproductive outcomes across different assays and clinical settingsAssayReferenceStudy overviewSDF groupingConclusion (by outcome detail)SCDLiu et al. (120)Assessed predictive value of different SDF levels in IVF-ET/ICSISDF ≤15%, 15%< SDF <30%, SDF ≥30%Fertilization rate, high-quality embryo rate, clinical pregnancy rate, cleavage rate: no significant differencesSCDSolanki et al. (121)Studied SDF in men with RPLSDF <30%, SDF ≥30%RPL: higher proportion of men with unexplained RPL had SDF ≥30%SCDWang et al. (8)Examined effect of SDF on IVF outcomes in unexplained infertilitySDF <25%, SDF ≥25%Live birth rate: lower. Clinical pregnancy rate, early miscarriage rate, fertilization rate, high-quality embryo rate: no significant differencesSCDZhang et al. (122)Investigated impact on IVF/ICSI outcomesSDF <25%, SDF ≥25%Fertilization rate, cleavage rate, high-quality embryo rate, implantation rate, clinical pregnancy rate, miscarriage rate, delivery rate, live birth rate: no significant correlationsTUNELFerrigno et al. (123)Studied SDF in morphologically normal sperm in ICSISDF <15%, SDF >15%Sperm chromatin: significantly higher percentage of morphologically normal sperm with chromatin alterations in patients with SDF ≥15%TUNELEsbert et al. (124)Analyzed factors influencing successful deliverySDF <36%, SDF ≥36%Miscarriage and successful delivery: using a 36% threshold had no impact on clinical outcomesAORex et al. (37)Studied pregnancy rates after IUISDF ≤10%, SDF >10%Pregnancy and live birth rates: significant differences between high/low SDF groups; SDF >10% detrimental to IUI cyclesAOYang et al. (38)Investigated impact on blastocyst formation and neonatal outcomesSDF< 15%, 15%≤ SDF <30%, SDF ≥30%Blastocyst formation rate, transferable embryo rate: significantly lower in high SDF group. Neonatal outcomes: associated with increased risk of low birth weight. Pregnancy outcomes, maternal adverse events: no significant differences among three groupsAOJiang et al. (39)Assessed impact on early embryo development and established thresholdsSDF ≤21.15%Fertilization rate: negative correlation. Embryological parameters: no associationPredictive value: SDF =21.15% showed low sensitivity and specificity for predicting high fertilization rateSCSAChristensen et al. (40)Studied ongoing pregnancy rates in IVF/ICSISDF >15%; SDF >25%Ongoing pregnancy (beyond 12 weeks): significantly reduced with SDF >15% (ICSI) and SDF >25% (IVF)SCSAKrog et al. (41)Investigated relationship between RPL and SDFMedian comparison; >25%Baseline SDF: no significant correlation. Couples not pregnant: 35.7% of men had SDF >25%SCSAChen et al. (42)Analyzed impact of different SDF levels on multiple IVF/ICSI outcomesSDF <15%, 15%< SDF ≤30%, SDF >30%Semen parameters, high-quality embryo rate, blastocyst development rate, pregnancy rate: all significantly reduced in high SDF group (>30%)SCSAZhu et al. (43)Assessed predictive value for IUI clinical outcomesSDF <30%, SDF≥30%Biochemical pregnancy rate, clinical pregnancy rate, miscarriage rate: no significant differences between groups. Predictive effectiveness: limited; highest sensitivity at SDF =16.75%SCSABest et al. (125)Predicted recurrent implantation failure (RIF) and pregnancy ratesMultiple groups (≤15% to ≥30.1%)Pregnancy odds, RIF risk, live birth, spontaneous miscarriage: SDF showed no predictive valueSCSALi et al. (44)Assessed predictive ability for IVF/ICSI pregnancy outcomesSDF ≤15%, 15%< SDF <30%, SDF ≥30%Pregnancy rate, RIF: neither SDF nor HDS could predict. Miscarriage rate: positive correlation with SDF. Birth weight: negative correlation with SDFSCSAZhu et al. (45)Evaluated impact of SDF on IVF-ET clinical outcomesSDF <30%, SDF ≥30%Delivery rate (fresh cycles), biochemical/clinical pregnancy rates (frozen cycles): significantly reducedSCSASaleh et al. (46)Investigated negative impact on IUI outcomesSDF <15%, 15%< SDF <24%, 25%< SDF<30%, SDF >30%No statistical analysis of correlation with IUI outcomes was reportedSCSAWang et al. (48)Examined effect on IVF outcomes in women with PCOSSDF ≤15%, SDF >15%High-quality blastocyst formation rate: lower. Fertilization rate, high-quality embryo rate, clinical pregnancy rate, miscarriage rate: no impactSCSAShen et al. (126)Assessed predictive ability for ART outcomes.SDF ≤15%, 15%< SDF <30%, SDF ≥30%Embryo implantation rate (fresh ICSI): significantly higher in normal SDF group. Live birth rate (fresh ICSI): significant difference between normal and abnormal SDF groupsSCSAM Bungum et al. (25)Assessed predictive value for IUI, IVF, and ICSI outcomesSDF ≤27%, SDF ≥27%IUI pregnancy/delivery rates: significantly higher in groups with SDF ≤27% and HDS ≤10%. IVF/ICSI outcomes: no statistical difference; ICSI performed betterSCSABungum et al. (24)Explored use of SDF and HDS in predicting ART outcomesSDF ≤30%, SDF ≥30%IUI pregnancy outcomes: significantly reduced in SDF >30% group. IVF/ICSI comparison: no difference when SDF ≤30%; when SDF >30%, ICSI outcomes were significantly better than IVFAO, acridine orange; ART, assisted reproductive technology; HDS, high DNA stainability; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF-ET, in vitro fertilization and embryo transfer; PCOS, polycystic ovary syndrome; RIF, recurrent implantation failure; RPL, recurrent pregnancy loss; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Summary of studies evaluating the association between SDF thresholds and reproductive outcomes across different assays and clinical settings AO, acridine orange; ART, assisted reproductive technology; HDS, high DNA stainability; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF-ET, in vitro fertilization and embryo transfer; PCOS, polycystic ovary syndrome; RIF, recurrent implantation failure; RPL, recurrent pregnancy loss; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; SDF, sperm DNA fragmentation; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Discussion Clinical applicability of SDF testing The clinical scenarios in which SDF testing provides actionable information, along with proposed intervention strategies, are summarized in (39,40,47,56,57,79,83,89,106-109,127-129) Table 8. The following subsections elaborate on the evidence supporting each scenario. Table 8Clinical scenarios for SDF testing and intervention strategiesClinical scenarioValue of SDF testingIntervention/management strategyUnexplained infertility20–50% of patients have abnormal SDF, suggesting a potential etiology (83,106) Threshold of 20% has best discriminatory power (89)Identify correctable factors (varicocele, lifestyle). If SDF improves post-intervention, consider natural conception/IUI. If ineffective or female window limited, consider ICSIVaricocele (127)Preoperative assessment of damage severity strengthens intervention necessity (35). Postoperative monitoring of intervention effect (108)Varicocele repair recommended for men with elevated SDF. Postoperative SDF reduction is a good prognostic factor. Persistent SDF elevation post-surgery warrants consideration of ICSI (108)RPLClearly associated with RPL (40,56,79,107,109), independent of female factors (RPL incidence up to 30% when SDF >15%) (128)Investigate male factors. Consider TESE + ICSI, antioxidants, donor sperm, varicocele repair, frequent ejaculation (evidence limited, requires more research)IUIPregnancy rates significantly reduced when SDF >25–30% (as low as 3%) (57)Prioritize identifying/correcting male factors. If SDF remains elevated post-intervention, consider ICSI (39). Testing valuable for patients about to start or who have failed IUIIVF/ICSIAffects pregnancy rates, embryo quality, miscarriage rates (22,108); patients with HDS ≥15% may require ICSIWhen SDF >30%, ICSI outcomes are superior to IVF. Testicular sperm ICSI is effective for patients with high SDF (improves pregnancy, live birth rates, lowers miscarriage rate)Birth defect riskHigh SDF may be associated with genetic abnormalities in offspring (47,129) (clear in animal studies (35), human evidence remains debated)Provide genetic counseling; weigh ART risks; special attention for advanced paternal ageART, assisted reproductive technology; HDS, high DNA stainability; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; RPL, recurrent pregnancy loss; SDF, sperm DNA fragmentation; TESE, testicular sperm extraction. Clinical scenarios for SDF testing and intervention strategies ART, assisted reproductive technology; HDS, high DNA stainability; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; RPL, recurrent pregnancy loss; SDF, sperm DNA fragmentation; TESE, testicular sperm extraction. Unexplained infertility Approximately 10–30% of infertile couples remain without a clear cause after routine clinical and laboratory evaluation (130,131). When basic assessments (physical exam, tubal patency, ovulation, semen analysis) are normal, it is termed “unexplained infertility” (132). When basic assessments (physical exam, tubal patency, ovulation, semen analysis) are normal, it is termed “unexplained infertility” (133,134). In men with idiopathic infertility, the proportion with abnormal SDF is as high as 40–50% (135,136). Studies show that an SDF threshold of 20% most effectively distinguishes infertile from fertile men (137). For such couples, SDF testing is recommended. An abnormal SDF result suggests that sperm chromatin damage may be the underlying cause. The male partner should be referred to a urologist/andrologist to identify and treat correctable factors (e.g., varicocele, lifestyle). After successfully reducing SDF, some couples may achieve natural conception or have improved ART outcomes. If SDF elevation cannot be corrected, especially when the female partner’s age or reproductive window is a concern, ICSI is recommended, as it is less affected by high SDF than IUI or conventional IVF. Using testicular sperm in non-azoospermia men with elevated SDF offers a new option for fertility treatment (138). Testicular sperm extraction (e.g., TESA, TESE) yields motile sperm with lower SDF, providing a potential strategy to improve ART outcomes in men with severe male factor infertility. Varicocele Varicocele is the most common correctable cause of male infertility (129). Its presence leads to venous stasis and oxidative stress, key mediators in the development of SDF and testicular dysfunction (139). Men with clinical varicocele (primarily grades 2 and 3) have elevated SDF rates (140,141). Varicocele repair is associated with a simultaneous reduction in oxidative stress markers and SDF (30). Therefore, SDF testing has value in varicocele management: preoperatively assessing the extent of damage to strengthen the case for intervention (129), and postoperatively monitoring the effect of intervention. A reduction in SDF is a good prognostic factor for natural conception and ART. Persistently abnormal SDF after surgery is a poor predictor and should prompt consideration of IVF/ICSI (30). RPL High SDF levels are positively correlated with miscarriage (70,87,113-116). SDF levels are elevated in couples with RPL, whether conceiving naturally or through ART (142). According to the latest AUA guidelines, DNA fragmentation testing should be considered for couples with unexplained RPL, and various treatment modalities, including TESE with ICSI, antioxidant therapy, donor sperm, varicocele repair, and/or frequent ejaculation, are employed. Currently, no well-controlled published studies assess whether any of these therapies reduce the risk of RPL (31). Intrauterine insemination Abnormal SDF negatively impacts Intrauterine Insemination (IUI) pregnancy rates. When the SCSA SDF index >30%, patients undergoing IUI have a significantly reduced predicted pregnancy rate, with the chance of conception potentially as low as 3% (110,143), and a 13% rate of spontaneous miscarriage. Additionally, a clinical threshold of 25% has been suggested for SDF in natural conception or IUI (44). If this threshold is exceeded, switching to ICSI fertilization in the ART lab is recommended. In contrast to the 25–30% thresholds commonly cited in the literature, a prospective cohort study by Rex et al. suggested that even a baseline DFI >10% was associated with significantly lower pregnancy rates and live birth rates following IUI (144). A clinical threshold of 25% has also been suggested for HDS. SDF testing is valuable not only for couples who have experienced unexplained IUI failure but also for those about to start treatment. If SDF is too high, priority should be given to identifying and managing male factors that may cause elevated SDF. After intervention and retesting, the decision to proceed with IUI can be made. If no clear causative factor is found, or SDF remains high post-intervention, switching to ICSI can be considered. IVF/ICSI Most IVF/ICSI meta-analyses agree that sperm DNA integrity influences reproductive success. Abnormal SDF affects IVF/ICSI pregnancy rates (30,127,145). Some studies suggest that elevated SDF is associated with reduced pregnancy rates in conventional IVF but not in ICSI (82,116,146). Clinically, there is preliminary evidence linking HDS with poor IVF fertilization rates; men with HDS ≥15% may require ICSI. Men with high SDF levels (≥30%) have a lower blastocyst rate and a higher risk of failing to achieve an ongoing pregnancy. SCSA provides valuable prognostic information for clinicians counselling couples before IVF and/or ICSI cycles (73). When no causative factor is identified or SDF remains elevated after treatment, ICSI using testicular sperm is considered an effective method to overcome unexplained ICSI failure (30,147). Notably, when sperm DNA damage is high, the embryo quality achieved with ICSI is often superior to that with conventional IVF. This difference may be because ICSI injects sperm directly into the oocyte within hours of ejaculation, potentially reducing iatrogenic damage that sperm might incur in the in vitro environment (148). Furthermore, a large-scale study suggested that IVF and ICSI may require different DFI thresholds for predicting pregnancy outcomes (15% vs. 25%, respectively) (149). Birth defects ART, while helping infertile couples achieve pregnancy, raises concerns about potential genetic risks—specifically, whether defective genomes might be passed to offspring. Current research conclusions are inconsistent. Lewis and Simon [2010] (62) noted that in men undergoing ICSI with severely elevated SDF, no direct inheritance of defective genes to offspring was observed. However, several studies suggest a direct or indirect association between the degree of sperm DNA damage in men and genetic abnormalities in offspring, including an increase in aneuploidy and other genomic abnormalities (150). Animal studies show that increased SDF in mice is associated with growth abnormalities, behavioural changes, and increased tumour rates in their offspring (129). Furthermore, older men with compromised sperm DNA integrity may have a higher risk of genetic diseases, birth defects, and neurological dysfunction in their children (151). Mechanisms of SDF impact on pregnancy outcomes The core mechanism of SDF’s impact on pregnancy outcomes lies in the integrity of the paternal genetic material and its dynamic balance with the oocyte’s repair capacity. When the level of sperm DNA damage exceeds the oocyte’s repair capacity, the zygote responds through non-apoptotic mechanisms, including slowing the replication rate of the paternal DNA and potentially causing chromosomal rearrangements, ultimately leading to embryonic developmental abnormalities, implantation failure, or spontaneous miscarriage (152-154). This long-term paternal effect may result from a combination of factors, including chromosomal aberrations, defects in cellular processes, or cell division delays. If paternal DNA damage cannot be effectively repaired, blastomeres may undergo apoptosis, leading to decreased embryo quality or developmental arrest (103). Most studies indicate that SDF levels are primarily associated with an early paternal effect. However, Tesarik et al. suggested that early paternal effects might not be related to SDF levels, as early paternal effects could still impair ART outcomes even without elevated SDF. The late paternal effect, however, is significantly correlated with increased DNA fragmentation, indicating that sperm DNA integrity mainly affects embryo morphological development after the zygote and early cleavage stages (119). Furthermore, studies have found that sperm DNA integrity is not the only paternal factor controlling early embryonic development. Paternal effects exist in the first cell cycle after ICSI, which may stem from genetic factors (e.g., gene activity in the male pronucleus) or epigenetic factors (e.g., sperm-derived oocyte activation factors or centrosome function) (92). The paternal genome may not become effective until the four- to eight-cell stage (155), and major qualitative changes occurring between the four- and eight-cell stages depend on transcription. These findings may explain the lack of correlation observed between SDF and fertilization rate, as well as between SDF and embryo quality (expressed as the “good” embryo rate). Female factors, particularly oocyte DNA repair capacity and maternal age, are integral to this mechanistic framework. Oocyte repair capacity is not static; it declines with advancing female age, reducing the oocyte’s ability to resolve sperm DNA lesions before embryonic genome activation (27,28). Consequently, the same level of SDF may result in different downstream molecular consequences, including persistent DNA breaks, chromosomal rearrangements, and delayed paternal replication, depending on the oocyte’s intrinsic repair machinery. This mechanistic interaction explains why the negative effects of high SDF on pregnancy outcomes have been reported to be more pronounced in studies involving older female populations. Thus, within the mechanistic model, female age and oocyte quality directly modulate the threshold of damage that can be tolerated, providing a biological basis for the variability observed across different clinical populations. Reported thresholds: a descriptive synthesis Currently, no unified standard exists for SDF thresholds in relation to pregnancy outcomes. Most studies support the association between high SDF and adverse pregnancy outcomes. The most frequently referenced threshold in SCSA and TUNEL studies is 30%, but this value did not predict poor outcomes in all studies. When SDF ≥30%, numerous studies report significant associations with decreased high-quality embryo rates, blastocyst development rates, and pregnancy rates. Studies have shown that the odds of achieving clinical pregnancy are seven times higher when SDF <30% (109). The threshold for reduced clinical pregnancy rates is often between 25% and 30% (156). However, a considerable number of studies show that SDF levels have no significant impact on fertilization rates, cleavage rates, clinical pregnancy rates, or live birth rates in different ART cycles (e.g., IVF/ICSI/FET), especially when lower thresholds (e.g., 15% or 20%) are used, sensitivity increases but predictive value remains limited (17,157-159). This inconsistency may stem from the non-uniform thresholds affecting results, and these conflicting results are a source of debate regarding whether SDF is a necessary test. Analysis of existing literature reveals significant variation in the thresholds used for SDF, with no established standard. The SDF cut-off values employed in studies range widely, from 10% to 36%, with common cut-offs including 10%, 15%, 20%, 25%, 27%, 30%, and 36%. Common cut-offs include 15%, 20%, 25%, and 30%. Grouping methods include binary (e.g., <15% vs. ≥15%; <30% vs. ≥30%), ternary (e.g., ≤15%, 15%< SDF <30%, ≥30%), and more detailed multi-category groupings (e.g., ≤15%, 15.1–20%, 20.1–25%, 25–30%, ≥30.1%). The ternary grouping (≤15%, 15%< SDF <30%, ≥30%) is most common. The reasons for the lack of a unified threshold include: first, methodological differences. Different detection methods (SCSA, TUNEL, SCD, Comet) have different underlying principles, making results difficult to compare directly. There is no established conversion formula between SCD and other methods (like SCSA), and although there is a significant positive correlation, numerical values may differ systematically. Second, varying threshold definition approaches. Some studies use binary classification, others use ternary or multi-category groupings, or analyze SDF as a continuous variable (160), making cross-study comparisons challenging. Third, study population heterogeneity. Populations differ by geography, age, infertility etiology, and the extent to which female factors are controlled. Fourth, outcome measure differences. Studies use various outcome measures (fertilization rate, high-quality embryo rate, clinical pregnancy rate, live birth rate, miscarriage rate), each with different sensitivity to SDF. Fifth, experimental condition influence. The threshold does not fully represent the absence of fertility. For example, a 30% threshold does not mean the other 70% of sperm have completely normal chromatin; it indicates that 30% of sperm exceeded the denaturation threshold under that specific experimental condition (87). Stronger DNA denaturation induction conditions would cause this proportion to fluctuate accordingly (161). To further examine these sources of heterogeneity, it is instructive to disaggregate the evidence by assay platform and by population subgroup. For SCSA, Evenson et al. (87) reported that no couple achieved pregnancy when COMPαt ≥ 30%, a level considered “not compatible with good fertility”. Bungum et al. (109) subsequently identified a DNA fragmentation index (DFI) threshold of 27% for IUI, above which the odds of clinical pregnancy dropped significantly [odds ratio (OR) =16, 95% CI: 1.9–137]; in the same study, ICSI outcomes were significantly better than IVF when DFI exceeded 27%. Virro et al. (73) found that men with DFI ≥30% were at risk for low blastocyst rates (<30%) and no ongoing pregnancies. For TUNEL, Duran et al. (106) reported that no IUI cycles resulted in pregnancy when >12% of sperm showed DNA fragmentation, whereas Henkel et al. (83) demonstrated that TUNEL-determined DNA fragmentation was predictive for pregnancy in IVF, with patients having high fragmentation showing significantly lower pregnancy rates (19.05% vs. 34.65%). For SCD, Muriel et al. (77) reported that when DFI exceeded 25%, the clinical pregnancy rate dropped from 48.7% to 29.6% (adjusted OR =2.15, 95% CI: 1.14–4.05); the same study also found that fertilization rate was inversely correlated with DNA fragmentation (r=−0.245, P=0.045) and that higher fragmentation was associated with slower embryo development and poorer morphology on day 6 (47.7% vs. 29.4%, P=0.044). For Comet, no consistent cut-off has been established, with parameters such as tail DNA percentage showing assay-specific variability without a validated clinical threshold (38,39). These assay-specific differences reflect fundamental differences in what each test measures—SCSA detects susceptibility to acid-induced DNA denaturation, TUNEL labels direct strand breaks, SCD evaluates chromatin dispersion after protein extraction, and Comet quantifies DNA migration under electrophoresis. As a result, a 30% value from SCSA and a 30% value from TUNEL are not interchangeable; direct cross-assay comparisons are discouraged in the literature, and no validated conversion formula exists between methods. Beyond assay platforms, disaggregation by population further illustrates the variability. Female age and ovarian reserve represent one of the most consistently reported modifiers. Esbert et al. (158) compared IVF outcomes using own versus donor oocytes and found that SDF negatively affected pregnancy rates only in cycles using own oocytes from older women, whereas in donor oocyte cycles—where oocyte quality and repair capacity are presumed to be superior—the impact of SDF was attenuated. This observation aligns with the mechanistic framework discussed in “Mechanisms of SDF impact on pregnancy outcomes” section: oocyte DNA repair capacity declines with maternal age, meaning that a given SDF level may be clinically significant in an older population but not in a younger one. Similarly, female factors such as polycystic ovary syndrome (PCOS) may influence how SDF affects outcomes; Wang et al. (162) reported that in PCOS patients undergoing IVF, the effect of SDF on clinical outcomes was less pronounced than in non-PCOS populations. Male infertility etiology also contributes to population-specific threshold behavior. Men with varicocele exhibit higher baseline SDF levels compared to fertile controls (128,129,140,141), and varicocele repair has been associated with significant SDF reduction (30); however, post-repair SDF thresholds that predict natural conception may differ from those in men without varicocele. In couples with RPL, McQueen et al. (67) demonstrated that the SDF threshold associated with increased miscarriage risk was approximately 15%—lower than the 30% threshold commonly cited in general ART populations—suggesting that RPL populations may require a more stringent cut-off due to the cumulative effect of even modest DNA damage on pregnancy maintenance. In men with unexplained infertility, who constitute 10–30% of infertile couples (130,131), the proportion with abnormal SDF has been reported to be as high as 40–50% (135,136), yet the threshold that best distinguishes fertile from infertile men in this subgroup (20%) (137) differs from those used in ART prediction. Treatment modality further affects threshold interpretation. In IUI, Duran et al. (106) found no pregnancies when SDF exceeded 12% (TUNEL), whereas Bungum et al. (109) established a 27% threshold for IUI using SCSA. In IVF/ICSI, some studies suggest that ICSI partially overcomes the negative effects of SDF on fertilization (82,116,146), but thresholds for miscarriage and live birth remain similar. Geographic and ethnic differences may also contribute to threshold variability; studies from different regions have reported varying SDF cut-offs for similar outcomes (157-159,163), and differences in lifestyle, environmental exposures, and genetic backgrounds may influence both baseline SDF levels and the biological response to DNA damage. Collectively, these assay-specific and population-specific observations further clarify the heterogeneity outlined above and underscore that a single threshold cannot adequately serve all clinical scenarios. A descriptive summary of reported assay‑specific threshold ranges is presented in Table 9. This table is intended only to reflect published data and does not constitute a recommendation for clinical use. Moreover, diagnostic, prognostic, and intervention thresholds represent different clinical concepts, and this distinction should not be overlooked. Diagnostic thresholds identify men with abnormal SDF relative to fertile populations (typically 15–20% for TUNEL, 20–25% for SCD, 20–30% for SCSA) (77,106,126,137,164). Prognostic thresholds predict ART outcomes but vary widely (10–36%) depending on the outcome and population (67,77,80,91,109,158). Intervention thresholds guide clinical decisions such as switching from IUI to IVF/ICSI or using testicular sperm (commonly 25–30%) (61,106,109,128). These three threshold types serve different purposes and are not interchangeable. Given that this review is primarily based on observational studies, and given the heterogeneity in the thresholds used across studies as well as the differential impact of SDF on various aspects of pregnancy outcomes, we are unable to propose specific diagnostic, prognostic, or intervention thresholds. Table 9Reported threshold ranges across different assaysAssayReported diagnostic range (vs. fertile)Reported prognostic range (for ART outcomes)Commonly cited values in the literatureKey references reporting these valuesSCSA20–25%25–30% (for pregnancy); >30% (for miscarriage)20%, 25%, 27%, 30%Evenson 1999 (87), Bungum 2004 (109), Virro 2004 (73)TUNEL15%15–20% (for fertilization); >20% (for miscarriage)15%, 20%Duran 2002 (106), Henkel 2004 (83), Sharma 2016 (126)SCD20%25% (for IUI); 20–25% (for IVF/ICSI)20%, 25%Muriel 2006 (77), Fernández 2005 (164)CometVariable (assay-specific)Tail (%DNA in tail)—no consistent cut-off reportedNot standardizedRibas-Maynou 2012 (38), Nicopoullos 2019 (39)ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Reported threshold ranges across different assays ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination; IVF, in vitro fertilization; SCD, sperm chromatin dispersion; SCSA, sperm chromatin structure assay; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling. Detection methods Regarding detection methods, the SCSA has been described as the most extensively validated assay, with robust predictive value for miscarriage in some report (87). for assessing SDF. Therefore, for infertile couples, SCSA-determined SDF levels are among the best predictors of a favorable pregnancy outcome. SCD and Comet assays are commonly employed and provide clinically useful information alongside SCSA, while TUNEL remains widely used and endorsed by AUA/ESHRE. SCD is cost‑effective but operator‑dependent, whereas Comet offers detailed strand‑break data at the cost of greater technical complexity. Beyond technical characteristics, the threshold values reported for TUNEL, SCD and Comet assays deserve specific attention given their distinct measurement principles. For TUNEL, which directly labels DNA strand breaks by incorporating labeled dUTP at free 3'-OH ends, reported thresholds are generally lower than those of SCSA, typically ranging from 15% to 20%. Duran et al. (106) reported that no IUI pregnancies occurred when TUNEL‑determined SDF exceeded 12%. Henkel et al. (83) demonstrated that TUNEL was predictive for pregnancy in IVF, with significantly lower pregnancy rates in the high‑fragmentation group (19.05% vs. 34.65%). Sharma et al. (126) standardized a flow‑cytometric TUNEL protocol and proposed a threshold of 20% for abnormal SDF. The generally lower threshold values reflect the method’s direct detection of strand breaks, and direct application of the 30% SCSA cut‑off to TUNEL is not appropriate. In addition to SCSA, the SCD assay is another widely used method for detecting SDF in clinical practice, and its clinical utility in predicting miscarriage and recurrent pregnancy loss has also been demonstrated in observational studies (165). For SCD, which evaluates chromatin dispersion after protein extraction, the commonly cited threshold for predicting ART outcomes ranges from 20% to 25%. Muriel et al. (77) found that when SCD‑determined DFI exceeded 25%, clinical pregnancy rates dropped from 48.7% to 29.6% (OR 2.15). Fernández et al. (164) demonstrated that SCD correlates significantly with SCSA (r=0.80, P<0.001), but numerical values differ between the two methods, indicating that SCSA‑derived thresholds are not directly transferable to SCD. For Comet, which uniquely quantifies DNA migration under electrophoresis and can distinguish single from double strand breaks, thresholds are highly protocol dependent. Simon et al. (24) reported a diagnostic threshold of 25% for male infertility and a markedly higher prognostic threshold of 52% (neat) and 42% (density gradient centrifuged) for failed pregnancy. Ribas-Maynou et al. (38) found Comet measured double strand breaks elevated in RPL, and Nicopoullos et al. (39) noted that no single cut‑off can be universally applied due to protocol variability. SCD and Comet thresholds are method and protocol‑dependent, and direct application of a 30% universal cut‑off across these assays is not supported. Taken together, the methodological and threshold differences among TUNEL, SCD, and Comet—alongside SCSA—raise fundamental challenges for cross‑assay comparability. Each assay measures a distinct aspect of sperm DNA integrity: SCSA assesses susceptibility to acid‑induced DNA denaturation; TUNEL directly labels strand breaks; SCD evaluates chromatin dispersion following protein extraction; and Comet quantifies DNA migration under electrophoresis, with the additional ability to distinguish single‑ from double‑strand breaks. These mechanistic differences mean that the same numerical value obtained from different assays does not represent the same biological phenomenon. Direct conversion between assays is not supported by the literature, and no validated formula exists to translate a DFI from one method to another. Furthermore, differences in sample processing (neat semen vs. density gradient centrifuged), fixation protocols, scoring methods (manual vs. flow cytometric), and cutoff determination approaches (ROC derived vs. fixed percentiles) contribute additional variability even within the same assay platform. These challenges underscore the need for assay specific reference ranges and standardized protocols, rather than reliance on a single universal cut off across all testing methods. In recent years, detection technology has been evolving, with increasing application of AI. Deep learning and other AI technologies have been used for SDF prediction, showing success rates exceeding 80% (166,167). Furthermore, multi-parameter combined diagnosis is a new trend. Combining sperm SDF with CASA parameters can build a more comprehensive diagnostic system. Studies show that CASA parameters (total motility, progressive motility, etc.) are significantly negatively correlated with SDF, and SDF data based on a 15% threshold show better predictive ability (168). This multi-parameter integration strategy not only improves diagnostic efficacy but may also help embryologists select sperm with better DNA integrity during ICSI cycles, thereby improving clinical outcomes. We recognize that SDF results should be interpreted in the context of the specific assay used, and that direct cross‑assay comparisons are discouraged; accordingly, we have included Table 9, which provides a descriptive summary of reported assay specific threshold ranges. Limitations and future directions Limitations Several limitations of this review and the underlying evidence base should be acknowledged. These include: (I) substantial methodological heterogeneity across studies in assays, thresholds, and outcome measures, which precludes meta‑analysis; (II) inadequate control for female factors (e.g., maternal age, oocyte quality) and male etiologies in many studies; (III) a predominance of observational cohort studies and retrospective analyses, with a notable scarcity of prospective randomized controlled trials (RCTs); (IV) potential publication bias favoring positive associations; (V) limited generalizability beyond tertiary care settings and specific ethnic populations. Future directions Based on identified gaps, future research should prioritize: (I) international standardization of SDF testing protocols and establishment of assay‑specific reference ranges through large multi‑center studies; (II) well-designed RCTs to evaluate whether SDF-guided interventions (e.g., varicocele repair, antioxidants, testicular sperm ICSI) improve live birth rates; (III) integration of multi-omics and AI-based models to enhance predictive accuracy; (IV) large long‑term cohort studies to clarify whether elevated SDF affects offspring health; (V) development of couple-based personalized thresholds that incorporate female age and oocyte repair capacity; and (VI) cost-effectiveness analyses to inform healthcare policy and resource allocation. Summary Summary of key findings This review synthesizes evidence from over 150 studies to provide a comprehensive overview of the association between SDF and pregnancy outcomes. The key findings are threefold: (I) high SDF is consistently associated with adverse reproductive outcomes in a stage-specific manner—significantly reducing success in natural conception and IUI, impairing blastocyst development in IVF, and primarily increasing miscarriage risk in ICSI; (II) the most frequently cited threshold of 30% (SCSA/TUNEL) is not universally applicable, with reported cut‑offs ranging from 10% to 36% depending on assay platform, population, and outcome measure, and no validated conversion formula exists between different detection methods; (III) SDF testing has demonstrated clinical utility in specific scenarios including unexplained infertility, varicocele, RPL, and ART counseling, with testicular sperm ICSI emerging as a potential option for men with persistently elevated SDF and repeated ART failure. In the context of clinical practice, these findings reinforce that SDF should be viewed as a complementary tool to conventional semen analysis rather than a replacement. For clinicians, SDF testing offers actionable information for counseling couples about prognosis and treatment selection—particularly in deciding between IUI and IVF/ICSI, or considering testicular sperm retrieval. However, the lack of standardized thresholds across assays means that results must be interpreted with caution and in the context of the specific test used, the clinical scenario, and female partner factors. For policy development and implementation, the current evidence supports the integration of SDF testing into fertility care pathways for select indications (unexplained infertility, RPL, varicocele, and prior to ART). However, widespread implementation is hampered by methodological heterogeneity and the absence of universally accepted reference ranges. Health systems and professional societies should prioritize establishing standardized testing protocols, reimbursable indications, and quality assurance programs to ensure equitable access and consistency in clinical decision‑making. For future research, as outlined in “Limitation” section, priorities include assay standardization, prospective interventional trials, multi‑omics integration, long‑term offspring health studies, personalized threshold development, and cost‑effectiveness analyses. Addressing these gaps will be essential to translate SDF from a research tool into a robust, evidence‑based component of routine fertility care. Conclusions SDF is a critical marker of male fertility, affecting fertilization, pregnancy, and blastocyst formation. However, no unified threshold has been established; reported cut‑offs vary widely (10–36%) depending on the assay method, study population, and outcome measure. The 30% cut‑off remains frequently cited, particularly in SCSA- and TUNEL-based studies, but its generalizability across assays and populations remains unvalidated. The core mechanism involves sperm DNA damage exceeding oocyte repair capacity, leading to embryo instability. The impact varies by conception method: high SDF reduces success in natural conception and IUI; in IVF, it lowers blastocyst quality; in ICSI, it mainly increases miscarriage risk. For high SDF patients, the use of testicular sperm for ICSI may be considered as an option, particularly in those with repeated ICSI failure. SCSA has been described as the most extensively validated assay, while AI-based tools offer future potential for rapid prediction. SDF testing should complement routine semen analysis, especially in unexplained infertility, RPL, and varicocele. Supplementary The article’s supplementary files as Acknowledgments None. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Footnotes Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0544/rc Funding: This research was financially supported by the National Science Foundation of China (Nos. 82371633 and 82571865), Beijing Municipal Natural Science Foundation (No. 7212134), the Fundamental Research Funds for the Central Universities: Peking University Clinical Scientist Program (No. BMU2023PYJ H012), Beijing Research Ward Excellence Program, BRWEP (No. BRWEP2024W094090107), Innovation and Transformation Project, Peking University Third Hospital (No. BYSYZHKC202514), the Clinical Key Project of Peking University Third Hospital (Clinical Research Category) (No. BYSYZD2025006), and the Clinical Cohort Construction Project C of Peking University Third Hospital (No. BYSYDL2024011). Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0544/coif). The authors have no conflicts of interest to declare. References Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0544/prf Supplementary Materials The article’s supplementary files as