Should cryopreserved epididymal or testicular sperm be recovered from obstructive azoospermic men for ICSI?

BJOG: an International Journal of Obstetrics and Gynaecology November 2004, Vol. 111, pp. 1289 1293 DOI: 10.1111/j.1471-0528.2004.00411.x Should cryopreserved epididymal or testicular sperm be recovered from obstructive azoospermic men for ICSI? M. Griffiths, a C.R. Kennedy, a J. Rai, a L. Wilson, a A.R.E. Blacklock, a S.D. Keay b Objective To determine the effect of the anatomical site of sperm recovery on intracytoplasmic sperm injection (ICSI) embryo implantation, pregnancy and live birth rates in couples with isolated obstructive azoospermia as the sole cause of infertility. Design Controlled, single centre, retrospective clinical study. Setting University Hospital, Centre for Reproductive Medicine. Sample One hundred and fifty-one cycles of ICSI were performed, using surgically recovered sperm, between August 1996 and March 2002. Methods The outcome of ICSI, with surgically recovered sperm, was compared between epididymal (Group E) and testicular (Group T) derived sperm. Inclusion was limited to couples undergoing their first treatment cycle, where female age was 39 years and a minimum of five oocytes were available for injection. Women with a history of ovarian surgery, ultrasonic evidence of polycystic ovaries, uterine anomalies or hydrosalpinx were excluded. Main outcome measures Clinical pregnancy, implantation and live birth rate. Results Forty-two of 151 cycles met the strict inclusion criteria. Groups E and T were comparable with respect to age, basal serum FSH, ovarian response; number of oocytes injected and number of embryos available and transferred. No difference existed between Groups E and T in implantation, clinical pregnancy or live birth rate (28.8% vs 25.8%, 42.9% vs 42.9% and 39.3% vs 42.9%, respectively). Conclusions Cryopreserved epididymal and testicular sperm, from men with obstructive azoospermia, appear equally effective in ICSI. Epididymal recovery should remain the method of first choice for obstructive azoospermic men but further study of sperm DNA damage rates in different testicular sites is required. INTRODUCTION Intracytoplasmic sperm injection (ICSI) of cryopreserved, surgically recovered, sperm has enabled azoospermic men to father children 1,2 using epididymal 3 or testicular derived 4 6 sperm. Epididymal recovery has been favoured as it is less invasive, 7 epididymal sperm have undergone greater maturation 8 and comparison of epididymal and testicular sperm, in mixed unselected groups of patients, 6,9 11 have demonstrated similar clinical outcomes. Higher sperm DNA damage levels are present in infertile men 12 and the extent of damage correlates with fecundity, 13,14 suggesting DNA damage adversely affects a Centre for Reproductive Medicine, University Hospitals Coventry and Warwickshire NHS Trust, UK b Department of Biological Sciences, University of Warwick, Coventry, UK Correspondence: Dr S. D. Keay, Centre for Reproductive Medicine, University Hospitals Coventry and Warwickshire NHS Trust, Clifford Bridge Road, Coventry, CV2 2DX, England, UK. D RCOG 2004 BJOG: an International Journal of Obstetrics and Gynaecology sperm competence. In vitro studies support this notion, as increasing sperm DNA damage levels are detrimental 15 17 to oocyte fertilisation, embryo development and implantation. 11 Therefore, the observation of significantly higher levels of DNA damage in epididymal compared with testicular sperm (25% vs 17%, respectively) raised concern that the anatomical site of sperm recovery may directly influence ICSI outcome 18 and that testicular recovery be preferred in men with obstructive azoospermia as it lessened the chance of obtaining sperm with compromised DNA. However, this carries an increased operative risk, including developing haematomata, compared with epididymal aspiration. Furthermore, as sperm DNA damage may only manifest at embryo implantation, 4 with less evidence of an impact on pronucleus formation, 18 the potential differences in sperm DNA quality from these two sites needs to be assessed in relation to embryo implantation and live birth rates following ICSI rather than fertilisation rates alone. Previous studies comparing ICSI outcome using surgically derived sperm, from different anatomical sites, have not been standardised for factors known to influence treatment outcome 6,9 11 such as advanced female age www.blackwellpublishing.com/bjog

1290 M. GRIFFITHS ET AL. and diminished ovarian reserve. 19 21 These factors should be controlled for when comparing outcomes as potential confounders other than the site of sperm recovery may have accounted for, or masked, differences reported in previous studies. Our hypothesis was that greater DNA damage in epididymal sperm may manifest as lower embryo implantation rates compared with testicular derived sperm. The specific aim of this study was to determine whether differences in embryo implantation, clinical pregnancy and live birth rates existed between epididymal and testicular derived sperm in obstructive azoospermia, where potential confounding factors, known to influence the outcome of ICSI, had effectively been eliminated. This would determine whether the more invasive testicular sperm recovery, as routine, on all obstructive azoospermic men was justified. METHODS ICSI cycles utilising epididymal or testicular sperm, performed at our centre between August 1996 and March 2002, were identified. Confounding factors known to affect the implanting potential of individual embryos (advanced female age, hydrosalpinx and uterine abnormality) were excluded. Inclusion was limited to: (1) a couple s first attempt, (2) men 50 years with normal testicular function and a diagnosis of obstructive azoospermia, thus providing sperm from a normally functioning testis (see paragraph on male assessment below) and (3) women 39 years with two ovaries, no history of ovarian endometriosis or ovarian surgery, no ultrasonic evidence of polycystic ovaries or uterine anomaly and no evidence of hydrosalpinges. A normal ovarian reserve was inferred by an early follicular phase (day two to day four) serum FSH 10 IU/L (within four months of commencing treatment). Additionally, poor responders (women producing 4 metaphase II [MII] oocytes at the time of ICSI) were excluded from analysis. Table 1. Patient inclusion criteria. Parameter Limit Rationale Treatment cycle First Avoid bias of previous treatment and inclusion of multiple cycles Female history 39 years old Exclude factors known No uterine anomaly No hydrosalpinges to adversely affect implantation Male history 50 years old Limit impact of abnormal spermatogenesis due to male ageing Obstructive azoospermia Normal testicular function Sperm Cryopreserved Standardised sperm processing No. of oocytes collected 5 Exclude poor responders The application of strict inclusion criteria ensured that the only variable was the site at which sperm were recovered. The criteria and explanation for their use are shown in Table 1. Two groups: epididymal sperm (Group E, n ¼ 28) and testicular sperm (Group T, n ¼ 14) were compared. All men were assessed by the same consultant Urologist after two semen analyses had shown azoospermia. 22 A relevant history was taken. Those men who had not undergone previous vasectomy underwent physical examination and further investigation by means of routine blood tests, endocrine profile, karyotype and, where appropriate, transrectal ultrasound of the prostate and seminal vesicles. If these results were shown to be normal [i.e. normal size testis (>4 cm length, >20 ml volume) and normal hormone profile (FSH 1 7 IU/L, LH 1 8 IU/L)] with no ultrasonically detected defect, obstructive azoospermia was diagnosed. The same consultant urologist carried out the subsequent sperm retrieval procedure. Sperm retrieval was performed as an outpatient procedure under local anaesthesia of the scrotum (Lignocaine 2%) and spermatic cord (Marcaine 0.5%). Sperm retrieval was achieved, for each patient, using percutaneous sperm aspiration, microepididymal sperm aspiration and testicular sperm extraction either singularly or in combination. The initial retrieval method of choice was dependent upon the patient s clinical details relating to the condition of the epididymis at the time of surgery. When testicular sperm extraction was performed, normal spermatogenesis was confirmed by histology. Percutaneous sperm aspiration was performed, as previously described. 23 Microepididymal sperm aspiration was performed, as for percutaneous sperm aspiration, after scrototomy and exposure of the epididymis. Testicular sperm extraction was performed as previously described. 6 Recovered epididymal fluid and testicular tissue was immersed into either HEPES buffered MEM (Medicult, UK) or HEPES buffered Earles medium (Medicult) for examination and preparation. If both epididymal fluid and testicular tissue were recovered, they were kept separate at all times. Testicular tissue was minced to release sperm from the seminiferous tubules, using frosted end glass microscope slides. Epididymal fluid and minced testicular tissue, suspended in buffer salt solution, was combined in a 1:1 ratio with Test yolk buffer cyropreservation media (Irvine Scientific California, USA). If both epididymal and testicular sperm were recovered for any one patient, these sperm were prepared and frozen separately. The sperm/cryopreservation buffer mixture was then loaded into a number of 1-mL cryovials (Camlab, UK) in volumes that ranged from 200 to 500 AL, depending on the total volume to be stored. The vials were then suspended in liquid nitrogen vapour for 15 minutes and then immersed into liquid nitrogen until required. Ovarian stimulation, oocyte retrieval, embryo transfer and assessment of pregnancy outcome were performed as

WHICH SPERM FOR ICSI IN OBSTRUCTIVE AZOOSPERMIA? 1291 Table 2. Aetiology of obstructive azoospermia and source of sperm for ICSI. Source of sperm injected at ICSI Epididymis (n ¼ 28) Testis (n ¼ 14) Azoospermia aetiology Vasectomy 12 7 CBAVD a 3 1 Obstruction 13 5 Retrograde ejaculation 0 1 a Congenital bilateral absence of vas deferens. previously described. 24 Suppression of pituitary gonadotrophin secretion with Naferelin (by nasal spray) was started in the mid-luteal phase of the preceding ovarian cycle. Ovarian suppression was confirmed by transvaginal ultrasound (no follicular activity >10 mm, endometrium <4 mm) and/or serum oestradiol (E 2 ) levels less than 100 pmol/l. Once ovarian suppression was confirmed, ovarian stimulation was started using a subcutaneous injection of follicle stimulating hormone (Metrodin High Purity, urofollitrophin, or Gonal-F, Serono Laboratories, Welwyn Garden City, UK) 150 IU. Ovarian ultrasound scans were performed on days 8 and 11 and human chorionic gonadotrophin (hcg) (Profasi; Serono Laboratories) was injected and oocyte retrieval arranged once three or more follicles achieved a diameter of at least 17 mm. ICSI was performed as previously described. 25 Cryopreserved sperm samples were used for ICSI. All oocytes for an individual patient were injected with sperm from the same recovery site (i.e. either epididymis or testis). In preparation for ICSI sperm samples were thawed at room temperature and then washed twice using HEPES buffered Earles medium (Medicult) at a centrifuge speed of 400 g. The resulting resuspended samples were then split into two with one-half incubated at room temperature and one-half incubated at 37 C. Motile sperm for ICSI were then recovered, approximately 3 hours after washing, from the sample containing the most motile sperm at that time. Table 3. Clinical and ICSI cycle characteristics. Values are presented as mean (SD) unless otherwise indicated. Epididymal sperm (n ¼ 28) Testicular sperm (n ¼ 14) Male age (years) 37.3 (6.9) 37.6 (6.0) Female age (years) 32.1 (3.6) 31.6 (4.0) Female basal FSH (IU/L) 6.6 (1.0) 6.5 (1.4) Female basal LH (IU/L) 5.7 (1.7) 5.2 (1.6) Female BMI (kg/m 2 ) 23.7 (2.3) 23.8 (2.4) No. metaphase II 11.1 (4.4) 9.7 (4.0) oocytes collected Fertilisation rate (%) a 75 [17 100] 67 [47 100] No. of embryos transferred a 2[2 3] 2[2 3] a Median value [range]. Table 4. Clinical outcome of ICSI with epididymal and testicular derived sperm. Clinical pregnancies were defined by ultrasound confirmation of an intrauterine gestational sac and fetal heart activity. ICSI cycles utilising epididymal sperm (Group E) were compared with those utilising testicular sperm (Group T). Clinical pregnancy rate per embryo transfer, embryo implantation rate and live birth rate per embryo transfer were compared using the m 2 test. Neonatal outcome was also recorded from hospital records. RESULTS Forty-two of 151 cycles of ICSI performed at our centre using surgically recovered spermatozoa between August 1996 and March 2002 met the inclusion criteria. Of these 42 cycles, 19 of the men had had a vasectomy, 4 suffered congenital bilateral absence of the vas deferens, 1 patient suffered retrograde ejaculation and the remaining 18 patients suffered a post-testicular obstruction of unknown aetiology. Table 2 shows the source of sperm used for ICSI according to male aetiology. Table 3 shows details of patient age, aetiology and ICSI outcome measures for the cycles studied. Group E (n ¼ 28) and Group T (n ¼ 14) were comparable with respect to factors known to affect the outcome of assisted conception treatment. Female age, basal serum FSH, basal serum LH, body mass index (BMI), male age, oocytes injected per patient, fertilisation rate and number of embryos transferred were similar in each group. There was no significant difference between Group E and Group T in clinical pregnancy rate (42.9% vs 42.9%), implantation rate per embryo transferred (28.8% vs 25.8%) or live birth rate (39.3% vs 42.9%), respectively (Table 4). There were no fetal anomalies detected in the study groups and the singleton live births weights were similar in the two groups (data not shown). DISCUSSION Epididymal sperm (n ¼ 28) Testicular sperm (n ¼ 14) Outcome measure Clinical pregnancy rate (%) 42.9 42.9 Live birth rate (%) 39.3 42.9 Implantation rate (%) 28.8 25.8 This study confirms previous reports 19,21,26 that excellent live birth rates, comparable to those using ejaculated sperm, can be achieved using cryopreserved surgically recovered sperm.

1292 M. GRIFFITHS ET AL. Our relatively small sample size reflects the strict selection criteria used. No difference, with respect to potential confounding factors for testicular function or ovarian response (basal serum FSH and oocyte number), existed between the groups and restriction of analysis to the first cycle of ICSI eliminated bias that may arise from previous treatment or the inclusion of multiple treatment cycles. The selection criteria used enabled us to study ICSI results based solely on sperm type, in essentially normal fertile couples. To our knowledge, this is the first time that careful elimination of potential confounding factors has been attempted to solely assess how the anatomical site of sperm recovery affects ICSI outcome in obstructive azoospermia. We observed no significant difference in implantation rate and live birth rate between the epididymal and testicular groups. Despite the limited power of this study to detect differences in clinical outcome, these findings, together with existing reports, 11,15 17 suggest sperm competence is broadly similar for surgically extracted epididymal and testicular sperm. Testicular sperm recovery has been recommended for all obstructive azoospermic patients based on higher epididymal sperm DNA damage levels. 18 A strong correlation between the level of sperm DNA damage and the resultant 15 17 embryo s ability to implant has been demonstrated. Thus, it might be expected that higher sperm DNA damage levels would adversely affect ICSI outcome, which is not supported by our findings. We were unable to assess sperm DNA damage directly and cannot confirm whether the two sperm populations differed in their level of DNA damage. Potential subtle differences in levels of DNA damage between the two groups, although not directly impacting on live birth rates, could potentially affect childhood development. Such a difference could only be detected by the careful follow up of infants. The variance in conclusions made between our study and Steele et al. s 18 may relate to the sperm populations used. Steele et al. 18 measured DNA integrity in a mixed population of motile viable sperm, immotile viable and non-viable sperm from the epididymis and testis. It has previously been reported that motile sperm have a greater nuclear DNA integrity than immotile sperm when separated from the same sample. 27 Degrading non-viable sperm would therefore have a larger degree of DNA damage. The proportion of viable and non-viable sperm would be different in the epididymis and testis, with the testis containing the greater proportion of viable DNA intact sperm, exactly as observed by Steele et al. 18 We exclusively selected motile epididymal and testicular sperm for ICSI, thus eliminating the likely impact of the proportion of non-viable sperm present in the sample. As ICSI minimises the potential for injecting non-viable DNA damaged sperm, it circumvents the need to recover sperm from the testis where there may be a higher proportion of viable DNA intact sperm. Further work is required to investigate the DNA integrity and aneuploidy rates in epididymal and testicular sperm and to determine whether sperm viability accounts for the previously reported differences in DNA damage rates. If DNA damage levels are truly higher in epididymal compared with testicular derived motile sperm, follow up studies of children born from ICSI with surgically recovered sperm should specifically record the anatomical site of the sperm used. During the course of this study, very high implantation rates (27.8%) were identified in the highly selected group of patients, where female factors had been eliminated. These data raise the potential for single embryo transfer in this group, thus minimising the risk of multiple pregnancy. We did not address surgical sperm recovery in nonobstructive azoospermic men but a recent report suggests a significantly worse outcome. 28 The importance of accounting for possible confounding female factors is as important as in obstructive azoospermia. Furthermore, a recent metaanalysis indicates that the cause of obstructive azoospermia may affect ICSI outcome. 29 A lower miscarriage rate was observed in acquired obstructive azoospermia compared with congenital bilateral absence of the vas deferens, and men with an infective aetiology have a worse outcome compared with a non-infective cause. This is important when considering future studies which should take into account the clinical indication for surgical sperm recovery in addition to the anatomical site. In conclusion, these data suggest that both epididymal and testicular sperm are effective for use in obstructive azoospermic patients undergoing ICSI, when female factors are eliminated, and questions whether routine, more invasive, testicular aspirations are necessary. DNA damage in sperm remains a legitimate concern however and warrants further evaluation before recommending testicular aspirations as first choice for obstructive azoospermic patients. References 1. Tournaye H, Camus M, Vandervorst M, et al. Surgical sperm retrieval for intracytoplasmic sperm injection. Int J Androl 1997;20:69 73. 2. Freidler S, Raziel A, Strassburger D, et al. Factors influencing the outcome of ICSI in patients with obstructive and non-obstructive azoospermia: a comparative study. Hum Reprod 2002;17:3114 3121. 3. Schlegel PN, Berkeley AS, Goldstein M, et al. Epididymal micropuncture with in vitro fertilization and oocyte micromanipulation for the treatment of unreconstructable obstructive azoospermia. Fertil Steril 1994;61:895 901. 4. De Croo I, Van der Elst J, Everaet K, De Sutter M, Dhont M. Fertilization pregnancy and implantation rates after ICSI with fresh or frozen thawed testicular spermatozoa. Hum Reprod 1998;13:1893 1897. 5. Devroey P, Liu J, Nagy Z, Tournaye H, Silber SJ, Van Steirteghem SJ. Normal fertilization of human oocytes after testicular sperm extraction and intracytoplasmic sperm injection. Fertil Steril 1994;62: 639 641. 6. Silber SJ, Van Steirteghem AC, Liu J, Nagy Z, Tournaye H, Devroey P. High fertilization and pregnancy rate after intracytoplasmic sperm injection with sperm obtained from testicle biopsy. Hum Reprod 1995;10:148 152.

WHICH SPERM FOR ICSI IN OBSTRUCTIVE AZOOSPERMIA? 1293 7. Schlegel PN, Su L-M. Physiological consequences of testicular sperm extraction. Hum Reprod 1997;12:1688 1692. 8. Robaire B, Pryor JL, Trasler JM. What does the epididymis do and how does it do it? In: Handbook of Andrology. Lawrence: Allen Press, 1995. 9. Dohle GR, Ramos L, Pieters MHE, Braat DDM, Weber RFA. Surgical sperm retrieval and intracytoplasmic sperm injection as a treatment of obstructive azoospermia. Hum Reprod 1998;13:620 623. 10. Ghazzawi IM, Sarraf MG, Taher MR, Khalifa FA. Comparison of the fertilising capability of spermatozoa from ejaculates epididymal aspirates and testicular biopsies using intracytoplasmic sperm injection. Hum Reprod 1998;13:348 352. 11. Balaban B, Urman B, Isiklar A, et al. Blastocyst transfer following intracytoplasmic injection of ejaculated, epididymal or testicular spermatozoa. Hum Reprod 2001;16:125 129. 12. Zini A, Fischer MA, Sharir S, Shayegan B, Phang D, Jarvi K. Prevalence of abnormal sperm DNA denaturation in fertile and infertile men. Urology 2002;60:1069 1072. 13. Duran EH, Morshedi M, Taylor S, Oehninger S. Sperm DNA quality predicts intrauterine insemination outcome: a prospective cohort study. Hum Reprod 2002;17:3122 3128. 14. Evenson DP, Larson KL, Jost LK. Sperm chromatin structure assay: its clinical use for detecting sperm DNA fragmentation in male infertility and comparisons with other techniques. J Androl 2002;23: 25 43. 15. Morris ID, Ilott S, Dixon L, Brison DR. The spectrum of DNA damage in human semen as assessed by single gel electrophoresis (Comet assay) and its relationship to fertilization and embryo development. Hum Reprod 2002;17:990 998. 16. Sakkas D, Manicardi G, Bizzaro D, Bianchi PG. Possible consequences of performing intracytoplasmic sperm injection (ICSI) with sperm possessing nuclear DNA damage. Hum Fertil 2000;3:26 30. 17. Ahmadi A, Ng S-C. Developmental capacity of damaged spermatozoa. Hum Reprod 1999;14:2279 2285. 18. Steele EK, McClure N, Maxwell RJ, Lewis SEM. A comparison of DNA damage in testicular and proximal epididymal spermatozoa in obstructive azoospermia. Mol Hum Reprod 1999;5:831 835. 19. Freidler S, Raziel A, Soffer Y, Strassburger D, Komarovsky D, Ron-El R. The outcome of intracytoplasmic injection of fresh and cryopreserved epididymal spermatozoa from patients with obstructive azoospermia a comparative study. Hum Reprod 1998;113: 1872 1877. 20. Keay SD, Liversedge NH, Mathur RS, Jenkins JM. Assisted conception following poor ovarian response to gonadotrophin stimulation. Br J Obstet Gynaecol 1997;104:521 527. 21. Levi AJ, Raynault MF, Bergh PA, Drews MR, Miller BT, Scott Jr RT. Reproductive outcome in patients with diminished ovarian reserve. Fertil Steril 2001;76:666 669. 22. World Health Organisation. WHO Laboratory Manual for the Examination of Human Semen and Sperm Cervical Mucus Interaction, 4th edition. Cambridge, UK: Cambridge University Press, 1999. 23. Meniru GI, Gorgy A, Batha S, Clarke RJ, Podsiadly BT, Craft IL. Studies of percutaneous epididymal sperm aspiration (PESA) and intracytoplasmic sperm injection. Hum Reprod Update 1998;4:57 71. 24. Bhattacharya S, Hamilton MPR, Shaaban M, et al. Conventional in-vitro fertilization versus intracytoplasmic sperm injection for the treatment of non-male-factor infertility: a randomised controlled trial. Lancet 2001;357:2075 2079. 25. Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 1992;340:17 18. 26. Tournaye H, Merdad T, Silber SJ, et al. No differences in outcome after intracytoplasmic sperm injection with fresh or with frozen thawed epididymal spermatozoa. Hum Reprod 1999;14: 90 95. 27. Kao SH, Choa HT, Wei YH. Multiple deletions of mitochondrial DNA are associated with the decline of motility and fertility of human spermatozoa. Mol Hum Reprod 1998;4:657 666. 28. Nicopoullos JDM, Gilling-Smith C, Almeida PA, Ramsay JWA. The results of 154 ICSI cycles using surgically retrieved sperm from azoospermic men. Hum Reprod 2004;19:579 585. 29. Nicopoullos JD, Gilling-Smith C, Ramsay JW. Does the cause of obstructive azoospermia affect the outcome of intracytoplasmic sperm injection: a meta-analysis. BJU Int 2004;93:1282 1286. Accepted 10 August 2004