A comparison between open and percutaneous needle biopsies in men with azoospermia

Human Reproduction vol.13 no.5 pp.1266 1271, 1998 A comparison between open and percutaneous needle biopsies in men with azoospermia B.Rosenlund 1,6, U.Kvist 3, L.Plöen 4, B.Lundh Rozell 2, P.Sjöblom 1 and T.Hillensjö 5 1 Department of Obstetrics and Gynaecology, 2 Department of Clinical Pathology and Cytology, Huddinge University Hospital, S-141 86 Huddinge, 3 Reproductive Medicine Centre, Department of Woman and Child Health, Karolinska Hospital, Stockholm, 4 Department of Anatomy and Histology, Swedish University of Agricultural Sciences, Uppsala and 5 Fertility Center Scandinavia, Göteborg, Sweden 6 To whom correspondence should be addressed Open testicular biopsy is a classic method of investigation in men with azoospermia. Recently, percutaneous needle biopsy of the testis has been used in attempts to obtain material for histopathological diagnosis in such cases and to retrieve spermatozoa for intracytoplasmic sperm injection (ICSI). To determine whether a 19 gauge (G) and a 21G butterfly needle could be used for percutaneous aspiration of testicular tissue to determine the presence of mature spermatids and assess spermatogenesis, 10 patients (16 testes) and 12 patients (17 testes) underwent 19G or 21G needle biopsy respectively, immediately followed by open testicular biopsy, with both procedures under local anaesthesia. Biopsy with each needle size was compared with open biopsy. With the 19G needle, in the 14 cases where material was obtained there was full agreement with open biopsy regarding the presence or absence of mature spermatozoa, whereas with the 21G needle only nine of the 13 biopsies yielding material were predictive in this respect. Each needle size correlated poorly with open biopsy regarding evaluation of spermatogenesis. We conclude that percutaneous biopsy with a 19G butterfly needle is a quick and reliable method for demonstrating spermatozoa for ICSI. But for a detailed histopathological diagnosis, however, the needle biopsies gave poor results, whereas the material from the open testicular biopsies was assessable. Key words: azoospermia/histopathological diagnosis/open testicular biopsy/percutaneous testicular biopsy/spermatogenesis Introduction Testicular biopsies have mainly been performed in azoospermic men with a normal testicular size and normal serum levels of follicle stimulating hormone (FSH) to find out whether the spermatogenesis is normal or impaired or whether established azoospermia can be attributed to occlusions in the efferent duct system. Two new lines in the investigation and management of the infertile man have broadened the indications for testicular biopsy. One is the new molecular approach to the understanding of how chromosome deletions (especially Y-derived) are related to spermatogenesis (Elliot et al., 1997). This necessitates a detailed evaluation of the process of spermatogenesis. A recently introduced therapeutic approach is to use testicular biopsy to obtain motile spermatozoa to treat infertility with intracytoplasmic sperm injection (ICSI) (Craft et al., 1993; Schoysman et al., 1993). In these cases men with small testicular volumes and men with elevated serum FSH levels have also come into consideration for testicular biopsy (Silber 1995; Tournaye et al., l996). Traditionally, open testis biopsy has been the standard procedure for obtaining testicular tissue for histopathological diagnosis. In the last decade percutaneous needle biopsy has been used as a diagnostic tool (Cohen and Warner, 1987; Rajfer and Binder, 1989) and also for testicular extraction of spermatozoa for ICSI (Bourne et al., 1995; Hovatta et al., 1995). There are no unequivocal results indicating the superiority of any particular technique open biopsy or different forms of needle biopsy. The aim of this study was to address the following two questions: Can testicular spermatozoa be aspirated with 21 gauge (G) or 19G butterfly needles? Can the different phases of spermatogenesis be evaluated in the aspirated tissue? Open testicular biopsies obtained immediately after needle aspiration served as controls. Materials and methods During the period December 1996 to April 1997 22 men, aged 23 to 48 years, underwent a total of 33 percutaneous testicular needle biopsies, immediately followed by an open biopsy as part of infertility investigation. All men had azoospermia and their testicular volumes, determined by orchidometry, ranged from 8 25 ml. No patient showed clinical signs of obstruction (distended epididymides or absence of vas deferens). During the first part of the study a 21G butterfly needle [21G3/4, 0.81 19 mm, Japan Medical Supply (S)PTE Ltd, Singapore] was used in 17 procedures (five bilateral and seven unilateral), group 21G, and in the second part of the study a 19G butterfly needle [19G3/4, 1.08 19 mm, Japan Medical Supply (S)PTE Ltd] was used in 16 procedures (six bilateral and four unilateral), group 19G. All biopsies were performed under local anaesthesia with a preoperative injection of morphine (Oxikodon-skopolamin-hydrobromid 8 mg 0.4 mg; Apoteksbolaget, Umeå, Sweden). The scrotal area was washed with 0.5% chlorhexidine solution followed by physiological saline and draped. Seven millilitres of prilocaine hydrochloride (Citanest 10 mg/ml; Astra, Södertälje, Sweden) was infiltrated around the vas deferens as described by Li et al. (1992), using a 22G, 5 cm needle (Microlance 3, 22G2 0.7 50 mm; Becton Dickinson, Dublin, Ireland). Additionally, 1 ml prilocaine hydrochloride was infiltrated into the scrotal skin. The procedure was repeated on the other side in patients who underwent bilateral biopsy. The testis was grasped between the thumb and forefinger of the non- 1266 European Society for Human Reproduction and Embryology

Percutaneous needle biopsies in azoospermia dominant hand. To prevent epididymal injury, the testis was displaced anteriorly. Two to four percutaneous punctures were performed with the butterfly needle in the upper pole of the testis. The needle was attached to a 20 ml (Plastipak ; Becton Dickinson, Drogheda, Ireland) syringe and a continuous negative pressure was applied during each puncture while the needle was being moved and directed in different areas. For the open biopsy an incision was then made over the same area immediately after the percutaneous biopsy. The size of the open surgery biopsy was about 3 3 6 mm. All patients left the clinic within 4hoftheoperation. Evaluation of the testicular biopsy specimens The biopsy specimens were fixed in 4% formaldehyde in isotonic phosphate buffer, ph 7.0, dehydrated and embedded in 2-hydroxyethyl methacrylate (Technovite 7100; Heraeus Kulzer GmbH, Friedrichsdorf, Germany). Sections, approximately 3 µm thick, were stained with haematoxylin and eosin and examined under the light microscope. Several scoring systems have been described in the literature, but none of these was suitable for the present study, in which the comparison between percutaneous needle biopsy and open surgery biopsy was the principal aim, together with an evaluation of the value of needle biopsies in predicting the presence or absence of testicular spermatozoa and/or of late, fully condensed spermatids. In each case we made an assessment of where, along the process, spermatogenesis was disrupted. Spermatogenesis can be subdivided into three phases: spermatocytogenesis, mitotic proliferation of diploid spermatogonia resulting in a species specific number of primary spermatocytes followed by a short premeiotic interphase, during which the last DNA synthesis takes place; meiosis, consisting of a long meiotic prophase and the two meiotic divisions, the first resulting in secondary spermatocytes, which almost immediately undergo the second meiotic division resulting in round spermatids; and spermiogenesis, the differentiation of a round spermatid into an elongated spermatid with condensed chromatin, which is released from the epithelium on spermiation. When the mature spermatid is shed at spermiation, it is called a testicular spermatozoon. The testicular spermatozoa flow with the seminiferous tubular fluid produced by the Sertoli cell via the rete testis and efferent ductules into the epididymal duct. The observations were classified as follows: disrupted spermatogenesis, where all developmental stages are seen but where spermatogenesis is quantitatively reduced, and degenerating cells of various developmental stages are present; disrupted spermatocytogenesis, where the production of spermatogonia, and hence of primary spermatocytes, is impaired; disrupted meiosis, where there are primary spermatocytes but spermatids are lacking. A high number of degenerated (necrotic) spermatocytes indicate that the meiotic divisions are not taking place, hence the lack of spermatids; disrupted spermiogenesis, where spermatids are formed, but differentiate abnormally or degenerate; complete absence of germ cells with only Sertoli cells present in the epithelium. The two extremes, normal spermatogenesis and complete absence of germ cells, often designated Sertoli-cell-only syndrome, are easily assessed, whereas disruption or partial disruption of one or more of the phases of spermatogenesis requires more material for correct interpretation. The presentation of the histopathological observations is kept as short as possible, and we have tried to classify all biopsies into a limited number of groups. When there were insufficient amounts of mechanically unaltered tissue in the specimens they were classified as unevaluable for a histopathological diagnosis. The study was approved by the local ethics committee. Results An overview of the results is presented in Tables I and II. In group 21G, the material from 16 of 17 open biopsies could be properly evaluated, but the remaining open biopsy was technically unsatisfactory (Table I). Among the needle biopsies, four yielded no material for embedding and none of the remaining 13 could be properly evaluated on account of poor technical specimen quality, mainly consisting of mechanical damage to the seminiferous epithelium. This was chiefly reflected in exfoliation of the seminiferous epithelium, with a total loss of topographical relations between the developing germ cells. In the open biopsy specimens, elongated spermatids and testicular spermatozoa were present in eight, and lacking in nine. Among the eight cases in which elongated spermatids and testicular spermatozoa were present, this observation was also made in two of the needle biopsies. In the remaining six, there was no material in two cases and in four no such cells were observed. In the nine open biopsy specimens lacking elongated spermatids and testicular spermatozoa, absence of such cells was also noted in the corresponding material obtained with the 21G needle. Group 19G comprised 16 pairs of biopsies from ten men (Figures 1 3). All open biopsies, but only seven of the needle biopsies, could be fully evaluated (Table II). In these seven the results did not differ from those obtained by open biopsies. In all needle biopsies yielding material for embedding (14 of 16), the presence or absence of elongated spermatids and testicular spermatozoa could be demonstrated and was in conformity with the findings in the open biopsies (Figure 2). All operations were performed on an out-patient basis under local anaesthesia and were well tolerated by the patients, who were discharged within 4 h. There were no clinically significant complications. Discussion New approaches in research, diagnosis and treatment of male infertility have prompted the development of new methods for testicular biopsies. Since the ICSI procedure is performed with a single spermatozoon (Palermo et al., 1992), attempts to obtain single spermatozoa by testicular biopsy may make treatment possible. The role of biopsy has thus been expanded to include patients with elevated FSH and small testes. Fertilization by a testicular spermatozoon was first reported by Craft et al. (1993) and Schoysman et al. (1993). Testicular spermatozoa for ICSI have also been retrieved by percutaneous needle biopsy in cases of obstructive azoospermia (Bourne et al., 1995; Hovatta et al., 1995), and also by open biopsies in nonobstructive azoospermia (Devroey et al., 1995). Even round spermatids extracted from testicular tissue have been used successfully for fertilization (Fishel et al., 1995; Tesarik et al., 1995). Thus, with a simple needle procedure the technical requirements for treatment are overcome, and to answer the question of whether motile elongated spermatids and testicular spermatozoa are present, simple needle biopsy has been advocated as the method of choice. If diagnostic measures are minimized, we may be deprived of the possibility of characterizing what we are actually treating. Men with severe 1267

B.Rosenlund et al. Table I. Results of the histopathological evaluation of spermatogenesis and observations regarding the presence of testicular spermatozoa or elongated spermatids in testicular biopsy specimens (from 12 men) obtained by open surgery and percutaneous needle aspiration with a 21 gauge butterfly needle Patient Testis Volume Histopathological evaluation of spermatogenesis Presence of testicular spermatozoa or elongated spermatids no. (ml) Open biopsy Needle biopsy Open biopsy Needle biopsy 21/1 right 25 not possible not possible yes yes 21/2 left 18 partial meiotic disruption not possible no no 21/3 right 20 no germ cells no material no - 21/4 right 8 no germ cells not possible no no 21/5 right 8 partial spermatogenic, complete not possible no no spermiogenic disruption left 10 partial spermatogenic disruption not possible yes no 21/6 left 12 no germ cells no material no - 21/7 right 10 no germ cells not possible no no left 12 no germ cells not possible no no 21/8 right 15 partial meiotic disruption not possible no no 21/9 right 20 partial spermatogenic disruption no material yes - left 20 partial spermatogenic disruption no material yes - 21/10 right 15 partial spermatogenic disruption not possible yes no left 15 partial spermatogenic disruption not possible yes yes 21/11 right 25 partial meiotic disruption not possible yes no 21/12 right 12 partial spermatogenic disruption not possible yes no left 12 spermatogenic disruption not possible no no Table II. Results of the histopathological evaluation of spermatogenesis and observations regarding the presence of testicular spermatozoa or elongated spermatids in testicular biopsy specimens (from 10 men) obtained by open surgery and percutaneous needle aspiration with a 19 gauge butterfly needle Patient Testis Volume Histopathological evaluation of spermatogenesis Presence of testicular spermatozoa or elongated spermatids no. (ml) Open biopsy Needle biopsy Open biopsy Needle biopsy 19/1 right 20 partial spermatogenic disruption no material yes - left 20 partial spermatogenic disruption no material yes - 19/2 right 12 no germ cells no germ cells no no left 12 no germ cells no germ cells no no 19/3 right 15 partial spermatogenic disruption not possible yes yes left 12 partial spermatogenic disruption partial spermatogenic yes yes disruption 19/4 right 12 no germ cells no germ cells no no 19/5 right 15 partial meiotic disruption not possible yes yes 19/6 right 25 partial spermatogenic disruption partial spermatogenic yes yes disruption left 25 partial spermatogenic disruption partial spermatogenic yes yes disruption 19/7 left 8 no germ cells no germ cells no no 19/8 left 10 partial spermatogenic disruption not possible yes yes 19/9 right 20 partial spermatogenic disruption not possible yes yes left 20 partial spermatogenic disruption not possible yes yes 19/10 right 20 partial spermatogenic disruption not possible yes yes left 20 partial spermatogenic disruption not possible yes yes infertility (azoospermia, severe oligospermia, small testes, elevated FSH), who are under consideration for ICSI, constitute a group in which genetic impairment is likely to be the cause of the disturbance (Peterlin et al., 1997; Tournaye et al., 1997), which could then be transmitted to the next generation. Thus it is important to relate genetic impairment to specific disruptions in the various phases of spermatogenesis in order to increase our knowledge about gene involvement in the regulation of the process of spermatogenesis. These men should be offered genetic counselling and the investigating andrologist should perform a proper evaluation of spermatogenesis. As shown in this study, an open biopsy yields a specimen of good quality for evaluation of the different phases of 1268 spermatogenesis, whereas this criterion was not fulfilled by percutaneous aspiration through a 19 or 21G butterfly needle. However, as a tool for evaluating the presence or absence of mature spermatids, the 19G butterfly needle was as good as the open biopsy, whereas the 21G needle was less reliable. The efficiency of a cutting needle biopsy for testicular histological diagnosis has previously been demonstrated by Kessaris et al. (1995), who found a 95% correlation between percutaneous and open biopsy regarding the histological diagnosis. The biopsies were carried out on 24 testes (19 patients) and the percutaneous biopsy was performed with an ASAP18 core biopsy system (18G with 17 mm notch), using one pass through the testis. In a study by Mallidis and Baker (1994) on 26 men (47 testes), in whom the histological diagnosis was

Percutaneous needle biopsies in azoospermia Figure 1. Pairs of micrographs at the same original magnifications (a, d: 40,b,e: 200 and c, f: 600 ), from the open (a c) and the percutaneous needle (d f) biopsy. In this patient the needle biopsy was of such good quality that it could be evaluated and spermatogenesis was characterized as partially disrupted. The high power micrographs (c and f) show tubules where all developmental stages were found. Note the elongated, condensed spermatids. Patient 19/6, right testis. Scale bar in a and d 0.5 mm, in b and e 100 µm and in c and f 30 µm. compared between open biopsy and tissue aspiration with a 20G, either full agreement (56%) or slight differences (38%) in the degree of hypospermatogenesis were found. It thus appears that the diameter of the needle is of importance for obtaining adequate material for histopathological diagnosis and for the identification of mature spermatids. It is also conceivable that the correlation between needle and open testis biopsy is related to the status of spermatogenesis and the consistency of tissue. In all our cases with germ cell aplasia, tissue obtained by 19G needle biopsies was adequate for histopathological evaluation. In the study by Kessaris et al. (1995), seven of 19 patients had spermatogenesis within normal limits. In our study all of the 22 patients had impaired spermatogenesis. In a recent study by Craft et al. (1997) percutaneous needle biopsies with a 19 or 21G butterfly needle Figure 2. Pairs of micrographs at the same original magnifications (a, d: 40,b,e: 200 and c, f: 600 ), from the open (a c) and the percutaneous needle (d f) biopsy. In this patient the needle biopsy was mechanically damaged and spermatogenesis could not be evaluated, whereas mature spermatids could be detected. Patient 19/3, right testis. Scale bar in a and d 0.5 mm, in b and e 100 µm, and in c and f 30 µm. were performed in 19 male patients with subfertility, and tissue adequate for histological assessment of spermatogenesis was obtained in all cases. However, concurrent open biopsy (controls?) was only performed in two cases. The reason for the discrepancy between their results and those in our study, where the same sized needles were used, is unclear, but it could be due to the use of different criteria for evaluation of the various phases of spermatogenesis. The biopsy technique may also have been different. In another recent study by Friedler et al. (1997) testicular sperm retrieval by percutaneous fine needle sperm aspiration was compared with testicular sperm extraction by open biopsy in men with non-obstructive azoospermia. Spermatozoa for ICSI were found in four patients of 37 (11%) by percutaneous needle aspiration and 16 cases (43%) by open testicular biopsy. In this study a 21G butterfly needle was used. Lately, the first case of a delivery following ICSI of mature live testicular sperm cells collected in a case of hypergonadotrophic azoospermia with maturation arrest was reported (Lewin et al., 1996). In this case spermatozoa were 1269

B.Rosenlund et al. Figure 3. Pairs of micrographs at the same original magnifications (a, d: 40,b,e: 200 and c, f: 600 ), from the open (a c) and the percutaneous needle (d f) biopsy. In this patient germ cells could not be detected in any of the biopsies. Patient 19/2, right testis. Scale bar in a and d 0.5 mm, in b and e 100 µm, and in c and f 30 µm. retrieved by using 21 23G butterfly needles. In this paper the authors emphasize that the fine needle aspiration technique may enable the operator to reach more testicular areas than an open testicular biopsy, and thus increase the chance of retrieving spermatozoa for ICSI. There has been a fear that percutaneous biopsy, being a blind procedure, may cause vascular injury. According to Jarow (1990), the medial and lateral parts of the upper pole are less likely to contain major branches of the testicular artery than other regions and it has therefore been recommended that percutaneous biopsies be performed close to these areas. Harrington et al. (1996) reported the occurrence of intratesticular bleeding, observed sonographically as a hypoechoic region in the testicular parenchyma, within 30 min after the biopsy in 7% (four of 58) percutaneous biopsies performed with an 18G, 15 cm long Microvasive biopsy needle system with a notch of 17 mm (two biopsy specimens from each testicle). In contrast, the proportion of open testis biopsies resulting in signs of intratesticular bleeding was 29% (10 of 34). It thus appears that the needle biopsy is associated with fewer complications than the open biopsy. To what extent the diameter of the needle and the use of aspiration are related to the intratesticular bleeding or damage is not known. It is conceiv- 1270 able, however, that bleeding would be more frequent with a needle of larger diameter. In a recent study by Schlegel and Su (1997), 64 patients were evaluated after open testicular biopsy for non-obstructive azoospermia with serial scrotal sonography, histological analyses and evaluation of the success of repeated sperm retrieval attempts. At 3 months after the open biopsy, 82% (14 of 17) of the patients had ultrasonographic abnormalities in the testis suggesting resolving inflammation or haematoma at the biopsy site. By 6 months, the acute changes had resolved leaving linear scars or calcifications. Repeat testicular biopsies were more likely to retrieve spermatozoa if the second attempt was performed more than 6 months after the initial biopsy (12 of 15), relative to those performed within 6 months (one of four), suggesting transient adverse physiological effects after testicular biopsies. Permanent devascularization of the testis may occur after testicular biopsies. Based on studies by Jarow (1991), Schegel and Su (1997) postulated that multiple testicular biopsies with multiple incisions in the tunica albuginea may result in interruption of a sufficient proportion of testicular arteries to devascularize the testis. Therefore multiple testicular biopsies should be avoided to reduce the risk of permanent ischaemic testicular injury. To conclude, percutaneous testicular biopsy with a 19G butterfly needle under local anaesthesia is a safe and reliable method for assessment of the presence of mature spermatids in men with azoospermia and can also be used in the treatment of azoospermia with testicular sperm extraction and ICSI. However, to increase our knowledge about spermatogenesis and obtain a basis for genetic counselling, in our opinion only open biopsies offer a sufficient quality for evaluation. Such knowledge is a prerequisite for building up new DNA-based tools for the identification of mutations affecting spermatogenesis. Acknowledgements The authors thank Kerstin Fallkvist and Eva Nilsson for skilful technical assistance. This study was supported by grants from the Swedish Medical Research Council. References Bourne, H., Watkins, W., Speirs, A. and Baker, H.W.G. (1995) Pregnancies after intracytoplasmic injection of sperm collected by fine needle biopsy of the testis. Fertil. Steril., 64, 433 436. Cohen, M.S. and Warner R.S. (1987) Needle biopsy of testes: a safe outpatient procedure. Urology, 29, 279 281. Craft, I., Bennett, V. and Nicholson, N. (1993) Fertilising ability of testicular spermatozoa. Lancet, 342, 864. Craft, I., Tsirigotis, M., Courtauld, E. and Farrer-Brown, G. (1997) Testicular needle aspiration as an alternative to biopsy for the assessment of spermatogenesis. Hum. Reprod., 12, 1483 1487. Devroey, P., Liu J., Nagy, Z. et al. (1995) Pregnancies after testicular sperm extraction and intracytoplasmic sperm injection in non-obstructive azoospermia. Hum. Reprod., 10, 1457 1460. Elliot, D.J., Millar, M.R., Oghene, K. et al. (1997) Expression of RBM in the nuclei of human germ cells is dependent on a critical region of the Y chromosome long arm. Proc. Natl. Acad. Sci. USA, 94, 3848 3853. Fishel, S., Green, S. and Bishop, M. (1995) Pregnancy after intracytoplasmic injection of spermatid. Lancet, 345, 1641 1642. Friedler, S., Raziel, A., Strassburger, D. et al. (1997) Testicular sperm retrieval by percutaneous fine needle sperm aspiration compared with testicular

Percutaneous needle biopsies in azoospermia sperm extraction by open biopsy in men with non-obstructive azoospermia. Hum. Reprod., 12, 1488 1493. Harrington, T.G., Schauer, D. and Gilbert, B. (1996) Percutaneous testis biopsy: an alternative to open testicular biopsy in the evaluation of the subfertile man. J. Urol., 156, 1647 1651. Hovatta, O., Moilanen, J., von Smitten, K. and Reima, I. (1995) Testicular needle biopsy, open biopsy, epididymal aspiration and intracytoplasmic sperm injection in obstructive azoospermia. Hum. Reprod., 10, 2595 2599. Jarow, J.P. (1990) Intratesticular arterial anatomy. J. Androl., 11, 255 259. Jarow, J.P. (1991) Clinical significance of intratesticular arterial anatomy. J. Urol., 145, 777 779. Kessaris, D.N., Wasserman, P. and Mellinger, B.C. (1995) Histopathological and cytopathological correlations of percutaneous testis biopsy and open testis biopsy in infertile men. J. Urol., 153, 1151 1155. Lewin, A., Weiss, D.B., Friedler, S. et al. (1996) Delivery following intracytoplasmic injection of mature sperm cells recovered by testicular fine needle aspiration in a case of hypergonadotropic azoospermia due to maturation arrest. Hum. Reprod., 11, 769 771. Li, P.S., Li, S., Schlegel, P.N. and Goldstein, M. (1992) External spermatic sheath injection for vasal nerve block. Urology, 39, 173 176. Mallidis, C. and Baker, H.W.G. (1994) Fine needle tissue aspiration biopsy of the testis. Fertil. Steril., 61, 367 375. Palermo, G., Joris, H., Devroey, P. and Van Steirteghem, A.C. (1992) Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet, 340, 17 18. Peterlin, B., Kunej, T., Zorn, B. et al. (1997) In Barratt, C., De. Jonge, C., Mortimer, D. and Parinaud, J. (eds), Genetics of Human Male Fertility. EDK, Paris, pp. 66 75. Rajfer, J. and Binder, S. (1989) Use of biopsy gun for transcutaneous testicular biopsies. J. Urol., 142, 1021 1022. Schlegel, P.N. and Su, L-M. (1997) Physiological consequences of testicular sperm extraction. Hum. Reprod., 12, 1688 1692. Schoysman, R., Vanderzwalmen, P., Nijs, M. et al. (1993) Pregnancy after fertilization with human testicular spermatozoa. Lancet, 342, 1237. Silber, S.J. (1995) What forms of male infertility are there left to cure? Hum. Reprod., 10, 503 504. Tesarik, J., Mendoza, C. and Testart, J. (1995) Viable embryos from injection of round spermatids into oocytes. N. Engl. J. Med., 333, 525. Tournaye, H., Liu, J., Nagy, P.Z. et al. (1996) Correlation between testicular histology and outcome after intracytoplasmic sperm injection using testicular spermatozoa. Hum. Reprod., 11, 127 132. Tournaye, H., Lissens, W., Liebaers, I. et al. (1997) Heritability of sterility: clinical implications. In Barratt, C., De Jonge, C., Mortimer, D. and Parinaud, J. (eds), Genetics of Human Male Fertility. EDK, Paris, pp. 123 144. Received on October 8, 1997; accepted on January 29, 1998 1271