If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Obstetrics and Gynaecology, University of British Columbia and the Children's and Women's Hospital and Health Centre of British Columbia, Vancouver, BC, Canada
Department of Obstetrics and Gynaecology, University of British Columbia and the Children's and Women's Hospital and Health Centre of British Columbia, Vancouver, BC, Canada
Department of Obstetrics and Gynaecology, University of British Columbia and the Children's and Women's Hospital and Health Centre of British Columbia, Vancouver, BC, Canada
Department of Obstetrics and Gynaecology, University of British Columbia and the Children's and Women's Hospital and Health Centre of British Columbia, Vancouver, BC, Canada
Sperm DNA fragmentation (sDF) has emerged as a valuable tool for evaluating male fertility, yet the relationship between DNA fragmentation in the male gamete and idiopathic recurrent pregnancy loss (RPL) remains a topic of ongoing debate. Hence, a meta-analysis was conducted of 12 prospective and 2 retrospective studies involving 530 men with a history of RPL who underwent sDF testing compared with 639 fertile control participants. The main outcome measures were sDF measured by comet assay, TdT (terminal deoxynucleotidyl transferase)-mediated dUDP nick-end labelling (TUNEL), sperm chromatin dispersion (SCD) or sperm chromatin structure assay. Overall, couples with a history of idiopathic RPL demonstrated higher levels of sDF than fertile couples (average mean difference 11.98, P < 0.001). Subgroup analysis demonstrated a similar average mean difference between the RPL and control groups using SCD compared with TUNEL, while mean paternal age and mean sperm motility in the RPL groups tested by meta-regression demonstrated no significant effect on the mean differences in sDF (P > 0.10). These results support the diagnostic value of sDF over standard semen analysis, as well as a possible paternally derived genetic origin of unexplained RPL. Further prospective studies are required to further assess the predictive utility of sDF for assessing couples with unexplained RPL.
). Although the exact definition of RPL remains a topic ongoing debate, it is commonly defined as either two or more failed clinical pregnancies as documented by ultrasonographic or histopathologic examination (
RPL is estimated to affect 2–5% of couples and has been attributed to a myriad of causes including metabolic disorders, cytogenetic abnormalities, antiphospholipid antibody syndrome and uterine anomalies (
). Although evidence-based treatments for RPL have drastically improved pregnancy outcomes, over half of all cases remain unexplained and may be attributed to more controversial aetiologies such as luteal phase deficiency, chronic endometritis or inherited thrombophilia (
). Given its heterogeneous pathophysiology, RPL has been established as a unique entity in relation to miscarriage and infertility, thereby requiring specific investigations and treatments (
). Although female factors remain the most well-defined and thoroughly studied aetiologies for RPL, the male factor contribution to RPL has been less thoroughly investigated; emerging evidence, however, suggests that 40–50% of cases may be attributable to this (
). It is most commonly associated with deficiencies in semen volume and quality, as defined by World Health Organization (WHO) parameters first established in the 1980s (
). Among the qualitative factors described, sperm motility (asthenozoospermia) and sperm morphology (teratozoospermia) are functional characteristics commonly associated with infertility (
). As a result, only male karyotype is required for RPL evaluation, while the association between qualitative semen parameters including sperm DNA fragmentation (sDF) and unexplained RPL remains unknown.
Through advances in molecular genetics, the integrity of sperm DNA has been shown to affect fertilization, subsequent embryo development, implantation and pregnancy (
). The aetiology of sDF is multifactorial and can be affected by exposure to environmental toxins, cigarette smoking, varicocele and increasing paternal age. Although a certain degree of DNA damage occurs naturally during sperm transport, and can be repaired within the oocyte cytoplasm upon insemination, adverse pregnancy outcomes can result when the damage exceeds the cytoplasm's capacity for repair (
These observations consequently inspired the development of several sperm fragmentation assays; among the most commonly used are TdT (terminal deoxynucleotidyl transferase)-mediated dUDP nick-end labelling (TUNEL), the comet assay, the sperm chromatin structure assay (SCSA) and the sperm chromatin dispersion (SCD) assay (
). Although each assay purportedly measures different forms of DNA damage, it has been generally accepted that sDF measurement provides a more accurate representation of male fecundity because it has lower biological variability than conventional semen studies (
Unfortunately, considerable inter-laboratory variability exists that can influence the reliability of certain assays. Although certain tests, such as the SCSA, have demonstrated respectable consistency with strict protocol adherence (
The Sperm Chromatin Structure Assay (SCSA((R))) and other sperm DNA fragmentation tests for evaluation of sperm nuclear DNA integrity as related to fertility.
), defining thresholds for assessing male factor subfertility using sDF assays nevertheless remains a significant challenge. Although a recent meta-analysis demonstrated an association between sperm morphology and recurrent RPL (
Sperm deoxyribonucleic acid fragmentation is increased in poor-quality semen samples and correlates with failed fertilization in intracytoplasmic sperm injection.
). As a result, the relative effect and contribution of sperm fragmentation on RPL remains unclear.
Several meta-analyses have demonstrated an association between high levels of sperm fragmentation and failure of recurrent IVF/intracytoplasmic sperm injection (ICSI) (
Whether sperm deoxyribonucleic acid fragmentation has an effect on pregnancy and miscarriage after in vitro fertilization/intracytoplasmic sperm injection: a systematic review and meta-analysis.
also demonstrated that elevated sDF was associated with an increased incidence of miscarriage. However, a review of studies that investigate the relationship between sDF and RPL after natural conception had yet to be undertaken. In order to improve our understanding of the male factor contribution to RPL, our systematic review and meta-analysis aims to investigate whether an association exists between sperm fragmentation and RPL.
Materials and methods
This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (
The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
The Ovid MEDLINE, EMBASE, Cochrane Library and PubMed databases were reviewed from inception until 15 March 2018. The following English-language subject headings and key words were searched: Sperm.mp. OR Exp Spermatozoa/ AND Exp DNA Fragmentation/ AND (recur* adj5 pregnanc* adj5 loss*).mp. OR Abortion, Spontaneous/ OR Miscarriage*.mp.
All original research articles, including randomized and non-randomized controlled trials, cohort studies and patient series, were included. All included studies reported a numerical value for sDF calculated from among the following assays: TUNEL, comet, SCD or SCSA. Additional studies were extracted from the references in the full-text articles. We also considered published abstracts from conferences.
Study selection and data extraction
We selected studies that evaluated sperm DNA damage in couples with a history of unexplained RPL. The primary outcomes of interest was sperm fragmentatio (sDF) as measured by the following assays: TdT (terminal deoxynucleotidyl transferase)-mediated dUDP nick-end labelling (TUNEL), sperm chromatin dispersion (SCD) or sperm chromatin structure assay. Secondary outcomes were sperm morphology and motility. For studies to be eligible, outcome data with an explicit mean/median sDF value (with appropriate confidence intervals or ranges) for both study and control groups were extracted in 2 × 2 tables. We also recorded sperm DNA assay type, number of previous pregnancies/miscarriages and paternal and maternal age. Importantly, eligible studies needed to explicitly include a complete description and workup of inclusion and exclusion criteria for both paternal and maternal factors (Supplemental Table 1).
Newcastle-Ottawa Quality Assessment Scales were used to evaluate the quality of the observational studies (
). Two reviewers (J. T. and O. T.) completed the quality assessment, and any disagreements about inclusion were resolved by consensus.
Statistical analysis
All analyses were carried out in R version 3.4.1 (2017-06-30; The R Foundation, Austria).
Several studies were excluded from quantitative comparison because of differences in study design. Specifically, the study by Venkatesh and colleagues (
Cytogenetic, Y chromosome microdeletion, sperm chromatin and oxidative stress analysis in male partners of couples experiencing recurrent spontaneous abortions.
Archives of gynecology and obstetrics.2011; 284: 1577-1584
) was excluded from quantitative analysis owing to differences in inclusion/exclusion criteria among the study groups, and Leach and co-workers’ study (
Sperm deoxyribonucleic acid fragmentation assessment in normozoospermic male partners of couples with unexplained recurrent pregnancy loss: a prospective study.
Semen quality and pregnancy loss in a contemporary cohort of couples recruited before conception: data from the Longitudinal Investigation of Fertility and the Environment (LIFE) Study.
) were included for quantitative comparison, and a random effects meta-analysis was employed to calculate the estimated average mean difference as implemented in the R package ‘metafor’ (
). Mean differences from studies were weighted by inverse variance, which gives greater weight to larger and more precise studies. Heterogeneity of the exposure effects was evaluated graphically using forest plots (
). To assess the influence of specific studies on the results and overall heterogeneity, a sensitivity analysis was performed by sequentially leaving each study out and comparing the model characteristics.
Subgroup analyses were also performed using different sperm fragmentation assays, TUNEL and SCD, and by comparing studies with differing inclusion criteria for the number of losses: two or more, and three or more. Meta-regression was used to test whether these variables accounted for the variability among studies, while mean paternal age and mean sperm motility in the RPL group were also tested for their effects on outcome. Maternal age was not tested as there were too few studies that included it.
Results
The search produced a total of 305 results: 58 from MEDLINE, 194 from EMBASE and 53 from PubMed. Following removal of duplicates, 224 remained and each title and abstract were reviewed by two reviewers. Subsequently, 97 full texts were selected for full review and an additional 83 excluded, leaving 14 studies for systematic review and 13 studies for quantitative analysis. A single study (
) was identified through a review of references from selected studies. Reasons for exclusion included: case reports, non-English-language articles, systematic reviews, studies that failed to screen and exclude cases of maternal infertility, those pertaining to pregnancy loss instead of RPL, and studies that did not report specific values for sDF, motility and morphology in the experimental and control groups. The PRISMA flow chart of the reviewed studies is shown in Figure 1. Two reviewers (J. T. and O. T.) independently searched and reviewed the retrieved articles, and the results were compared. Any disagreement was resolved by discussion.
Figure 1Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram showing the search for studies. SFI, (Sperm Fragmentation Index).
The main characteristics of the 14 studies are presented in Table 1. As shown in Supplemental Table 2, the studies scored well according to Newcastle-Ottawa Quality Assessment Scales: one study scored 9 of 9 points, six studies scored 8, seven studies scored 7, and one study scored 5. Most importantly, all the included studies provided clear definitions of RPL as either two or more, or three or more previous pregnancy losses, and screened for appropriate male and female factors through detailed reproductive history and infertility evaluation, karyotype analysis and appropriate metabolic and immunological screening. Maternal anatomical factors were screened by pelvic ultrasound, hysterosalpingogram or sonohysterogram and appropriately excluded if an abnormality was found. Males with chronic medical conditions, obesity and significant smoking or alcohol histories were also excluded.
Sperm deoxyribonucleic acid fragmentation assessment in normozoospermic male partners of couples with unexplained recurrent pregnancy loss: a prospective study.
Semen quality and pregnancy loss in a contemporary cohort of couples recruited before conception: data from the Longitudinal Investigation of Fertility and the Environment (LIFE) Study.
Cytogenetic, Y chromosome microdeletion, sperm chromatin and oxidative stress analysis in male partners of couples experiencing recurrent spontaneous abortions.
Archives of gynecology and obstetrics.2011; 284: 1577-1584
As shown in Figure 2, the random effects model revealed a significant association between RPL and sperm fragmentation (estimated average mean difference 11.98, 95% confidence interval [CI] 6.64–17.32, P < 0.001), suggesting that men in the RPL groups had higher average percentages of sperm fragmentation than those in the control groups. There was significant heterogeneity between studies (I2 = 97.3%, Q test P < 0.0001), which can be seen in the forest plot, where studies do not show overlapping confidence intervals. There was no evidence of systematic bias in the funnel plot (Supplemental Figure 1), and Egger's test was non-significant (P = 0.82). Leave-one-out sensitivity analysis suggested that the removal of two studies – by Khadem and colleagues (
Sperm deoxyribonucleic acid fragmentation assessment in normozoospermic male partners of couples with unexplained recurrent pregnancy loss: a prospective study.
) – might lead to modest reductions in tau-squared values from 89.5 to 77.1 and 71.7, respectively. However, their exclusion makes very little difference to the heterogeneity as measured by I2 (being over 96% in both situations), or to the estimated mean difference (10.8 and 10.6, respectively, compared with11.98); therefore there is no evidence to suggest that they are overly influential in the final analysis. The sDF and sperm parameter results for each study included in the meta-analysis are included in Table 2.
Figure 2Forest plot of association between sperm DNA fragmentation and recurrent pregnancy loss. CI, confidence interval; df, degrees of freedom.
Sperm deoxyribonucleic acid fragmentation assessment in normozoospermic male partners of couples with unexplained recurrent pregnancy loss: a prospective study.
Semen quality and pregnancy loss in a contemporary cohort of couples recruited before conception: data from the Longitudinal Investigation of Fertility and the Environment (LIFE) Study.
Cytogenetic, Y chromosome microdeletion, sperm chromatin and oxidative stress analysis in male partners of couples experiencing recurrent spontaneous abortions.
Archives of gynecology and obstetrics.2011; 284: 1577-1584
Subgroup analysis by sperm fragmentation assay type suggested that the average mean difference between the RPL and control groups was similar between the two groups of studies, and the amount of heterogeneity between the studies was high and similar (mean difference in SCD 11.17, 95% CI 0.73-21.61, P=0.002) Figure 2).
Subgroup analysis by inclusion criterion of the number of losses suggested that the average mean difference was very similar between the two groups of studies, with high heterogeneity (mean difference for two or more losses 13.56, 95% CI 5.52–21.61, P = 0.0009; mean difference for three or more losses 11.95, 95% CI 3.72–20.18, P = 0.004; Supplemental Figure 2a). Furthermore, subgroup analyses between studies that explicitly excluded cases of varicocele and/or a smoking habit demonstrated a consistently significant average mean difference between the RPL and control groups (average mean difference 14.56, 95% CI 7.50–21.62, P < 0.0001; Supplemental Figure 2b). Finally, as shown in Supplemental Figure 3, neither paternal age nor sperm motility demonstrateda significant relationship with mean difference in sDF when evaluated by meta-regression (P > 0.10).
Discussion
Although previous meta-analyses assessed the association between high levels of sperm fragmentation on both miscarriage rates (
Whether sperm deoxyribonucleic acid fragmentation has an effect on pregnancy and miscarriage after in vitro fertilization/intracytoplasmic sperm injection: a systematic review and meta-analysis.
), this is the first systematic review and meta-analysis to investigate the relationship between sperm fragmentation and RPL, which is considered to be a related, yet unique, aetiological entity (
). Despite the fact that the male factor contribution to unexplained RPL has largely been poorly studied, it is a well-known contributor to adverse pregnancy outcomes and will probably increase in prevalence as various studies have demonstrated a global decline in overall sperm quality and increased incidence of male factor subfertility (
). Ultimately, this study assessed whether an association between sDF and RPL exists independent of more conventional semen parameters such as sperm motility.
Our meta-analysis demonstrates a clear association between a high frequency of sperm DNA damage and RPL compared with fertile control couples. In the absence of major maternal factors, this positive association suggests the involvement of a male factor in the pathogenesis of RPL and supports the diagnostic capabilities of sDF as a screening tool among patients with RPL. As Venkatesh and colleagues (
Cytogenetic, Y chromosome microdeletion, sperm chromatin and oxidative stress analysis in male partners of couples experiencing recurrent spontaneous abortions.
Archives of gynecology and obstetrics.2011; 284: 1577-1584
) have demonstrated in a prospective case control study, sDF was significantly elevated compared with a fertile control cohort despite normal morphology and motility, which demonstrates an independent association between sDF and RPL even where there is a normal semen analysis. Carlini and colleagues (
) also found that increased sperm DNA damage was associated with an increase in the number of spontaneous abortions, suggesting that greater DNA damage may increase the likelihood of impaired embryonic development. Finally, a recent observational prospective cohort study demonstrated that couples experiencing a pregnancy loss were more likely to have male partners with abnormal sDF (
Semen quality and pregnancy loss in a contemporary cohort of couples recruited before conception: data from the Longitudinal Investigation of Fertility and the Environment (LIFE) Study.
), the positive association demonstrated in this meta-analysis suggests that an elevated sDF reflects a high degree of sperm damage in couples with RPL that may be beyond the repair capacity of the oocyte, thereby giving rise to genetic mutations that are incompatible with post-implantation embryonic development and ongoing pregnancy. Notably, several recent studies have also demonstrated that sperm aneuploidy may be an important cause of elevated sDF, thereby supporting a potential role for preimplantation genetic testing in cases of unexplained RPL (
). In this way, sDF may hold significant diagnostic capability, and the elevated sperm fragmentation found in men of couples with RPL suggests the involvement of a male factor in the pathogenesis of RPL.
Furthermore, our subgroup analysis comparing different sperm fragmentation assays revealed a similar association with RPL using TUNEL or SCD. In a related meta-analysis, Robinson et al. demonstrated the strongest association among all assays tested as being between TUNEL and miscarriage (
) and the differences observed were attributed to the aspects of DNA damage tested by each of these techniques. For instance, TUNEL directly quantifies DNA damage through incorporation of labelled dUTP into single- and double-stranded DNA breaks at neutral pH and thereby does not require denaturation (
). Conversely, SCD requires a denaturation step and indirectly measures fragmented DNA by measuring differences in the halos of dispersed DNA loops that are produced by non-fragmented DNA following acid denaturation and removal of nuclear proteins (
). Nevertheless, both assays demonstrated similar association and variability in our study. Hence, clinicians can expect similar sensitivity and overall reliability for TUNEL and SCD when evaluating the contributing male factor in cases of RPL.
Ultimately, the additional information provided by sperm fragmentation testing over semen analysis may be beneficial for both diagnostic and treatment purposes. For example, high levels of DNA fragmentation on SCSA are predictive of poor outcomes after intrauterine insemination in couples with mild male factor infertility (
) also demonstrated that ICSI may be a preferred method of in-vitro treatment in cases of high sDF, noting that the odds of live birth per ovum retrieval was significantly lower with standard IVF if there was sDF of over 20% as measured by SCSA. Furthermore, the role of reactive oxygen species, a known cause of elevated sperm fragmentation (
), in the pathogenesis of RPL may be further investigated, as may the potential benefits of antioxidant therapy. Indeed, a recent Cochrane Review demonstrated the benefits of antioxidants to improve pregnancy rates among infertile couples after both natural conception and assisted reproductive technology, although the benefits specifically in cases of RPL were not addressed (
In the current meta-analysis, we encountered significant heterogeneity among the included studies. Although this probably reflects some variation in study characteristics, it is also important to note the inherent heterogeneity of patients with RPL. As demonstrated in other recent reviews on the topic (
), a myriad of underlying aetiologies that contribute to RPL make comparisons between study populations particularly challenging. Importantly, however, all the included studies excluded anatomical contributors to sperm fragmentation, such as varicocele, as well as several other factors such as body mass index, environmental exposures, caffeine and alcohol consumption, and smoking (Supplemental Table 1). Details of semen collection were reported in most studies, but differences in the duration of abstinence prior to semen collection as well as inter-laboratory variance could partially account for the heterogeneity observed in sperm fragmentation values between studies. Additionally, the definition of RPL varied between studies, although the majority defined RPL as two or more consecutive miscarriages. Inclusion and exclusion criteria for the female partner were clearly stated in most studies, although the number and gestational age of prior pregnancy losses were rarely defined. Both 2 retrospective and 12 prospective studies were included and a variety of different assays were used, which compounds this heterogeneity.
Furthermore, the results of our meta-regression demonstrate that the observed association between sDF and RPL occurs independent of differences in paternal age and sperm motility; however, the samples sizes were small for regression analyses, and since these are means of patient-level data and are not study-level variables, their effects may not actually represent the effects within studies (
). For example, no difference may be found between studies because of these variables, but this does not mean that these variables do not influence the results within studies. Without individual-level data, this issue is impossible to resolve.
Finally, it is important to note that although an average mean difference of 11.98% was statistically significant in this meta-analysis, thereby supporting the diagnostic capability of sDF in cases of idiopathic RPL, its predictive capacity cannot be inferred. Ultimately, the clinical benefit of such a diagnostic test can only be reliably implemented by first determining a standardized and validated threshold for defining abnormal DNA fragmentation in the context of RPL. Among the studies included, threshold values ranged from 18% (
) depending on the type of assay and protocols used, which highlights a major limitation of sDF as a clinical tool in much the same way as standard semen analysis (
High variability in results of semen analysis in andrology laboratories in Tuscany (Italy): the experience of an external quality control (EQC) programme.
) proposed a threshold of 20% to define abnormal sDF with high sensitivity and specificity, but this was based primarily on studies that evaluated infertile couples undergoing IVF/ICSI. Among couples with a history of idiopathic RPL after natural conception, Eisenberg and colleagues (
Semen quality and pregnancy loss in a contemporary cohort of couples recruited before conception: data from the Longitudinal Investigation of Fertility and the Environment (LIFE) Study.
) prospectively examined the relationship between sDF and RPL and found a clear association; however, the authors noted that further corroboration and larger prospective randomized studies were required to assess the definitive predictive capacity of sDF for informing future risk of RPL. Ultimately, a single meaningful threshold value of sDF across all assay types is probably not possible given the intrinsic differences in assay technique (
The Sperm Chromatin Structure Assay (SCSA((R))) and other sperm DNA fragmentation tests for evaluation of sperm nuclear DNA integrity as related to fertility.
Based on this systematic review and meta-analysis investigating the association between sDF and idiopathic RPL, our results demonstrate an overall strong association between high levels of sperm fragmentation and RPL using both TUNEL and SCD assays. As genetic abnormalities are thought to contribute to early miscarriage, our findings support the notion of a possible paternally derived genetic origin of recurrent unexplained pregnancy loss. Further prospective studies are still required to assess the diagnostic and predictive strength of sDF for assessing couples with unexplained RPL.
Supplemental Figure 3. Meta-regression evaluating the association between (a) average sperm motility and (b) average paternal age on the mean difference in the meta-analysis.
Sperm deoxyribonucleic acid fragmentation assessment in normozoospermic male partners of couples with unexplained recurrent pregnancy loss: a prospective study.
Semen quality and pregnancy loss in a contemporary cohort of couples recruited before conception: data from the Longitudinal Investigation of Fertility and the Environment (LIFE) Study.
The Sperm Chromatin Structure Assay (SCSA((R))) and other sperm DNA fragmentation tests for evaluation of sperm nuclear DNA integrity as related to fertility.
High variability in results of semen analysis in andrology laboratories in Tuscany (Italy): the experience of an external quality control (EQC) programme.
The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
Sperm deoxyribonucleic acid fragmentation is increased in poor-quality semen samples and correlates with failed fertilization in intracytoplasmic sperm injection.
Cytogenetic, Y chromosome microdeletion, sperm chromatin and oxidative stress analysis in male partners of couples experiencing recurrent spontaneous abortions.
Archives of gynecology and obstetrics.2011; 284: 1577-1584
Whether sperm deoxyribonucleic acid fragmentation has an effect on pregnancy and miscarriage after in vitro fertilization/intracytoplasmic sperm injection: a systematic review and meta-analysis.
Justin Tan is an Obstetrics and Gynecology Resident at the University of British Columbia, Canada. He received his Bachelors in Engineering from the Massachusetts Institute of Technology (MIT), USA, followed by a Masters in Public Health and Biostatistics from the University of Cambridge, UK, and most recently his medical degree from McGill University, Canada.
Key Message
Couples with a history of idiopathic recurrent pregnancy loss (RPL) demonstrate higher levels of sperm DNA fragmentation (sDF) than fertile couples. These results support the diagnostic value of sDF over semen analysis in the context of RPL and provide evidence of a possible paternally derived genetic origin of unexplained RPL.
Article info
Publication history
Published online: December 22, 2018
Accepted:
December 10,
2018
Received in revised form:
October 18,
2018
Received:
August 2,
2018
Declaration: The authors report no financial or commercial conflicts of interest.
Justin Tan is an Obstetrics and Gynecology Resident at the University of British Columbia, Canada. He received his Bachelors in Engineering from the Massachusetts Institute of Technology (MIT), USA, followed by a Masters in Public Health and Biostatistics from the University of Cambridge, UK, and most recently his medical degree from McGill University, Canada.
30658-8&id=gr1.jpg){kind=link}
30658-8&id=gr2.jpg){kind=link}