Retrospective multicentre study on mechanical and enzymatic preparation of fresh and cryopreserved testicular biopsies

V Baukloh on behalf of the German Society for Human Reproductive Biology


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Isolation of sperm suitable for ICSI from fresh or frozen–thawed testicular sperm extraction (TESE) can be facilitated by mechanical or enzymatic processing of the samples. METHODS: A retrospective multicentre study was initiated to compare these two approaches. Eleven German centres provided data on their TESE cycles performed during the period 1996/1997. Quality of retrieved sperm, fertilization rates of injected oocytes, embryo quality, resulting pregnancy rates and evolution of pregnancies were evaluated. RESULTS: The percentage of cycles with at least some motile sperm available for injection was higher after mechanical preparation. Independent of the preparation method, fertilization rates were higher for motile compared with immotile sperm or elongated spermatids in all groups and in general higher for cryopreserved versus fresh samples. Embryo quality was significantly better after injection of motile sperm for all treatments and in particular after enzymatic versus mechanical processing of biopsies. Pregnancy rates were identical for embryos derived from sperm prepared mechanically or enzymatically from fresh or cryopreserved testicular samples. The abortion rate (32/172, 18.6%) and the rate of multiple implantations (32/140, 22.9%) were not different from results reported in the literature for ICSI using ejaculated sperm. CONCLUSION: In this retrospective multicentre study, no unequivocal advantage of one over the other preparation method could be identified in 839 ICSI cycles using testicular sperm from 549 patients.

Key words: embryo quality/fertilization rates/pregnancy rates/sperm preparation/testicular biopsy


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The second most important advance in the treatment of severe male subfertility after the introduction of ICSI in 1992 (Palermo et al., 1992Go) was the isolation of sperm or spermatids directly from testicular biopsies (testicular sperm extraction; TESE) to be used as a sperm source in men with obstructive and non-obstructive azoospermia (Devroey et al., 1995Go; Nagy et al., 1995Go; Van Steirteghem et al., 1998Go; Nijs and Van Der Elst, 2000Go). The concept of cryopreserving several specimens from one biopsy for use in later treatment cycles was introduced in 1996 (Salzbrunn et al.) and now is applied widely in most of the centres offering this kind of assisted reproductive technology. The optimal method of obtaining suitable sperm for injection from testicular tissue is still under debate. Some laboratories use mechanical preparation by mincing and shredding the whole tissue (Verheyen et al., 1995Go). Others employ enzymatic digestion using collagenase (Salzbrunn et al., 1996Go; Crabbe et al., 1997Go) to loosen the cellular contacts in the tubular walls facilitating release of gametes.

In 1998 the German Society for Human Reproductive Biology decided during their annual general meeting to initiate a retrospective analysis of the results after TESE-ICSI comparing the data from groups applying mechanical or enzymatic preparation of testicular sperm from fresh or frozen–thawed biopsies. The intention of this study was to find out if one of the two methods of processing testicular tissue has any specific advantage over the other in terms of quality of recovered sperm, fertilization results, embryo quality or established pregnancies. The present paper summarizes the results of the compiled data of 11 centres covering the treatment period from 1996 to 1997.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
A questionnaire was sent out to all centres willing to participate in the detailed analysis of all TESE cycles performed during 1996–1997. The parameters asked were: female patient's age at treatment, number of oocytes retrieved, number injected and fertilized normally (two pronuclei = 2PN), number and morphological quality of embryos transferred [i.e. degree of embryo fragmentation as defined by the German National IVF Registry, [Deutches IVF Register (D.I.R.), 1997]: A, 0–4; B, 5–25; C, 26–50; D, >50%]; type of testicular biopsy (fresh or frozen–thawed), type of tissue preparation (mechanical or enzymatic processing), number and quality of testicular sperm used for injection (motile or immotile sperm, elongated spermatids; there were no cases of round spermatid injection); pregnancy rates and outcome of pregnancy (abortion, elective termination, birth).

Eleven centres contributed to the results presented. Complete data on all parameters mentioned above were available from nine centres (Table IGo). Centre 2 (20 cycles) did not report on patients' age at treatment, centre 6 (51 cycles) gave no details on number of patients treated, quality of isolated sperm and of resulting embryos. The number of cycles provided by each centre is shown in Table IIGo, however, the order in Table IIGo is different from Table IGo to ensure anonymity for each centre. The cycle results from individual centres were grouped according to the source of the testicular biopsy (fresh or cryopreserved; F or C) and the preparation method of tissue (mechanical or enzymatic; M or E) used for every specific treatment cycle. Statistical analysis was performed by {chi}2-test or analysis of variance and a statistical difference of P < 0.05 was considered significant. The number of cycles in the four subgroups and the contribution of data by the different centres were unevenly distributed: Fresh–enzymatic (F–E) 34 cycles: centre 2, 14; centre 4, 20; fresh–mechanical (F–M) 95 cycles: Centre 3, 60; centre 5, 11; centre 10, 24; cryopreserved–enzymatic (C–E) 486 cycles: Centre 1, 388; centre 2, 6; centre 4, 6; centre 7, 22; centre 8, 14; centre 9, 36; centre 11, 14; cryopreserved–mechanical (C–M): 224 cycles: Centre 3, 21; centre 5, 154; centre 10, 49.


View this table:
[in this window]
[in a new window]
 
Table I. Participating centres listed according to alphabetical order of locations (cities)
 

View this table:
[in this window]
[in a new window]
 
Table II. Number of cycles contributed by each of the centres
 
This, together with the retrospective type of the study, makes final conclusions difficult. Nevertheless the number of cases collected represent a large body of experience of different centres and may give some clues for clinical decisions.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The contribution of data and individual pregnancy rates of the 11 centres are listed in Table IIGo. There was considerable variation in pregnancy rates achieved by individual centres which, however, did not reach statistical significance (P = 0.1104) due to large differences in the number of treatment cycles. The overall pregnancy rate per transfer after TESE–ICSI was 20.9% which corresponds well with the ICSI results of 23.3% using ejaculated sperm reported for 1997 by the German National IVF Registry (D.I.R.).

Patient characteristics in the four subgroups
Some of the general parameters were significantly different between the four subgroups formed according to the source and preparation method of the testicular biopsy material (F–E, F–M, C–E, C–M; Table IIIGo). The average age of the females was significantly lower for patients represented in the F–E group and higher in the C–M group. Also the number of mature oocytes (metaphase II) available for ICSI was significantly different between groups. The lower yield of oocytes in the group with the oldest patients and the higher in the younger patients reflects the well known age-dependency of oocyte recovery rates (Silber et al., 1997Go).


View this table:
[in this window]
[in a new window]
 
Table III. Differences in patients' characteristics in the four subgroups
 
Quality of retrieved testicular sperm
Ten of the 11 centres reported details on the quality of testicular sperm after the respective preparation method (Table IVGo). The proportion of cases with no sperm available for injection of mature oocytes was quite low with 1.7% (14/839) in total. The percentage of cases with availability of at least some motile sperm ranged from 55.5% (centre 3) to 100% (centre 11) with no statistical differences due to the high variation in numbers of cycles treated per centre. Therefore it seems unlikely that the distribution of non-obstructive and obstructive azoospermia cases differed significantly between centres.


View this table:
[in this window]
[in a new window]
 
Table IV. Quality of testicular sperm available for injection of oocytes after preparation in individual cycles
 
In total the percentage of cycles with motile sperm was higher after mechanical (213/319 cases, 66.8%) versus enzymatic preparation (308/520 cases, 59.2%; P = 0.0347) and similar after using fresh compared with cryopreserved biopsies (86/129 cases, 66.7%, and 435/710 cases, 61.3%; P = not significant).

Fertilization rates obtained with testicular sperm
The results after injection of individual oocytes with motile and immotile sperm or elongated spermatids from fresh or cryopreserved biopsies are shown in Figure 1Go. The total fertilization rate was higher using cryopreserved compared with fresh sperm (C: 47.6 and 46.5%, versus F: 32.1 and 42.6%; {chi}2 = 35.3072, df = 3, P < 0.0001). In general the results after enzymatic and mechanical preparation were similar with some remarkable exceptions. For all motile sperm mechanical preparation gave higher fertilization rates than enzymatic treatment (M: 54.4 and 57.5% versus E: 34.7 and 52.5%; {chi}2 = 50.2125, df = 3, P < 0.00001) whereas for the subgroups using cryopreserved specimens (C–E and C–M) results were better with enzymatic preparation versus mechanical processing for both immotile sperm (38.5 and 21.2%; {chi}2 = 255.5280, df = 1, P < 0.00001) and elongated spermatids (2.1 versus 18.0%; {chi}2 = 62.7763, df = 3, P < 0.00001).




View larger version (80K):
[in this window]
[in a new window]
 
Figure 1. Normal fertilization rates of mature oocytes (percentage two pronuclei) according to preparation and quality of testicular sperm (motile, immotile, elongated spermatids) used for injection.

 
Quality of embryos resulting from ICSI with testicular sperm
Embryo quality judged before transfer with regard to degree of fragmentation is presented in Table VGo. The percentage of embryos derived from the injection of motile sperm differed considerably within the four subgroups (F–E: 91.4, F–M: 74.1, C–E: 68.1 and C–M: 89.3%; {chi}2 = 76.4013, df = 3, P < 0.00001). In all four subgroups embryo quality was scored better after injection of motile compared with immotile sperm or elongated spermatids. Comparison of the total number of embryos from all subgroups revealed a significantly higher percentage of grade A plus B, i.e. good quality embryos after enzymatic versus mechanical preparation (F–E: 93.8 and C–E: 78.6%, versus F–M: 57.1 and C–M: 63.0%; {chi}2 = 80.4994, df = 3, P < 0.00001). There was no statistical difference between subgroups F–M and C–M with regard to embryo quality while significantly more good quality embryos were seen in subgroup F–E than in subgroup C–E (P = 0.0009), although the difference in the size of subgroups was considerable.


View this table:
[in this window]
[in a new window]
 
Table V. Embryo quality as judged by degree fragmentation according to preparation of testicular sperm used for ICSI
 
Pregnancies after TESE–ICSI
The pregnancy rates obtained for the individual subgroups are depicted in Figure 2Go. The total success rates did not show significant differences (F–E: 18.8, F–M: 19.8, C–E: 19.4 and C–M: 22.1%; {chi}2 = 0.6730, df = 3, P = not significant). Likewise the delivery rates per transfer reported (data not shown) ranged from 14.3 to 19.4% and were not significantly different between the four subgroups ({chi}2 = 0.5913, P = not significant).



View larger version (37K):
[in this window]
[in a new window]
 
Figure 2. Pregnancy rates per transfer (%) according to sourceand preparation as well as quality of testicular sperm. Spermorigin: F = fresh; C = cryopreserved; E = enzymatic biopsy;M = mechanical preparation. Sperm quality = motile (only motile sperm used); motile/immotile (both motile and immotile sperm used); immotile (only immotile sperm used); immotile/elong. (both immotile sperm and elongated spermatids used); elong. spermatids (only elongated spermatids used)

 
More important than the source and preparation method was the quality of the testicular sperm used for injection of oocytes. In all combined cases where exclusively motile sperm were available the pregnancy rate was 26.1% per transfer (136/522); when both motile and immotile sperm had to be used this rate dropped to 15.0% (9/60). Corresponding rates for immotile sperm, mixed injections of immotile sperm and elongated spermatids, and elongated-spermatids-only cases were 5.9, 9.1 and 2.3% respectively (11/188, 1/11 and 1/44). These differences are statistically highly significant ({chi}2 = 47.0224, df = 4, P < 0.00001).

Evolution of pregnancies
All 11 participating centres provided data on the evolution of pregnancies achieved after ICSI using testicular sperm during the period 1996–1997 (Table VIGo). Of the 172 pregnancies 32 ended as spontaneous abortions (1/6 in F–E, 1/22 in F–M, 22/99 in C–E, 8/44 in C–M), 108 (62.8%) continued as intact singleton gestations and 29 sets of twins (one in F–E, five in F–M, 13 in C–E, 10 in C–M), as well as three triplets (one in C–E, two in C–M) were reported giving a multiple rate of 22.9% for the ongoing pregnancies. No major malformations were reported for the babies born after TESE–ICSI in this study.


View this table:
[in this window]
[in a new window]
 
Table VI. Evolution of pregnancies (n = 172) obtained after ICSI with testicular sperm (data from 11 centres)
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Differences between centres
Data on the number of obstructive and non-obstructive azoospermia cases within the cycles evaluated in this study were not available but the distribution of cycles with no sperm available for injection did not suggest that one of the two situations was over-represented in any of the centres. However, there was a tendency for accumulation of ‘unfavourable cases’ in centres 1, 2, 3 and 5 where, in a relatively high proportion of cycles (224/654, 34.3%), all available oocytes had to be injected with immotile sperm or elongated spermatids compared with the other six centres (4, 7, 8, 9, 10 and 11) which only had 9.7% (18/185 cycles) of such situations. Nevertheless the transfer and pregnancy rates between these two groups of centres were almost identical (transfer rates: 587/654, 89.8% and 169/185, 91.4%; pregnancy rates: 126/587, 21.5% and 32/169, 18.9%; no statistically significant differences).

Centres 1, 2, 3 and 5 also treated the majority of cycles in this survey (654 of 839, 77.9%; Table IIGo) which points to the fact that difficult cases obviously tended to be referred to experienced centres. The largest contribution to the total data set was by centre 1 with 388 cycles. These also represented 79.8% of the cases in subgroup C–E for the detailed analysis. Therefore the results for this subgroup were strongly influenced by the experience of this one centre and findings should be interpreted with caution.

A considerable bias was also probably introduced by the fact that individual centres in general applied only one preparation method (mechanical or enzymatic) so that comparisons may reflect differences between centres rather than between methods.

Comparison of fresh and cryopreserved testicular biopsy
The data shown indicate no major differences between the results with fresh or cryopreserved testicular tissue as the sperm source regarding implantation and pregnancy rates achieved (Figure 2Go) which is in agreement with findings of other investigators (Friedler et al., 1997Go; Habermann et al., 2000Go; Huang et al., 2000Go). The rate of oocyte fertilization appeared to be even higher after using sperm from cryopreserved than from fresh specimen (Figure 1Go) indicating that neither the process of cryopreservation itself nor the cryo-storage negatively influenced the fertilizing capacity of the testicular sperm. A similar observation was reported (Palermo et al., 1999Go) when comparing ICSI results using testicular sperm from fresh and frozen–thawed biopsies. Pregnancy rates achieved were the same but a higher fertilization rate with frozen–thawed than with fresh sperm was obtained (74.4 versus 60.4%) although this difference did not reach statistical significance due to the low number of cases (n = 5) included in the test group. The difference in fertilization rates observed in the present study cannot be clearly attributed to the source of the testicular sperm alone but was probably also influenced by variations between participating centres. Centres 2 and 4 performed a few cycles using fresh (n = 32) and a few using cryopreserved testicular sperm (n = 14) after enzymatic preparation, making a direct comparison possible. The fertilization rates of injected oocytes were identical under both conditions (121/346 and 42/120). Likewise, centres 3, 5 and 10 provided some data for comparison of injection results with fresh and with cryopreserved sperm after mechanical preparation (89 and 230 cycles) and also produced similar fertilization rates (337/750 and 762/1681). Therefore a final conclusion on the fertilizing potential of fresh compared with frozen–thawed testicular sperm cannot be drawn from the data presented here. A larger series of cases with both sperm sources applied in parallel and performed by only a few centres would be desirable to clarify this question.

All of the centres participating in this study applied cryopreservation of testicular biopsies at least for later treatment cycles. The advantages of this approach are now generally accepted. It allows the freedom of planning the oocyte retrieval independently from the biopsy in the male partner. In addition and more importantly the chances for successful injection of theoocytes can be predicted from the outcome of histological and preparative evaluation of the testicular biopsy material before treating the female partner with ovarian stimulation (Fischer et al., 1996Go; Küpker et al., 2000Go). For the male side one major advantage of thorough histological investigation is the possibility to detect carcinoma in situ very early and to refer affected patients to appropriate treatment before they engage in infertility treatment. This situation has been reported to occur in ~1% of infertility patients investigated (Schulze et al., 1997Go).

Comparison of preparation methods for testicular sperm
Both the enzymatic and the mechanical preparation of testicular sperm yielded high rates of motile sperm (55.5–100%) and these gave the highest fertilization rates after injection of mature oocytes (Figure 1Go) in all subgroups. Whether the higher proportion of motile sperm observed after mechanical versus enzymatic preparation was in fact due to the preparation method itself or was the result of a bias due to the very different group sizes cannot be clarified from the data available.

One important observation made in this study was a higher fertilization rate for immotile sperm and elongated spermatids after enzymatic compared with mechanical preparation of the tissue for the cryopreserved specimen (Figure 1Go; subgroup C–E versus C–M). This may be explained by easier processing of the injected sperm within the oocytes subsequent to membrane damage probably induced by the cryopreservation procedures. A similar mechanism was suggested byNagy et al. (1995) who attributed their high fertilization rate after ICSI with ejaculated sperm partly to damages of the sperm membrane caused by relatively high speed centrifugation during preparation (Nagy et al., 1995Go). Mechanical modification of the sperm membrane permeability as induced by touching the flagellum with the ICSI pipette also has been thought to improve the rate of sperm–oocyte interaction thus enhancing the fertilization rate (Nijs and Van der Elst, 2000Go).

Embryo quality as judged by the degree of fragmentation (Table VGo) was better in those embryos derived from the injection of motile compared with immotile sperm or elongated spermatids in all subgroups. The proportion of good quality embryos was superior after enzymatic compared with mechanical preparation but resulting pregnancy rates per transfer (Figure 2Go) were nevertheless almost identical between the four subgroups.

General comments
To reach a definite conclusion on the effect of the processing technique in specifically difficult cases (mainly immotile sperm or only elongated spermatids) the application of both preparation methods on aliquots from the same biopsy and injection of sibling oocytes would be necessary, which is not feasible in a clinical setting. The mechanical preparation approach has the advantage of being quite fast, taking only about 15 min, while the enzymatic technique is more time consuming requiring at least 4 h of preparation. A practicable approach based on the findings presented in this study could be to apply mechanical processing to specimens with a fairly good prognosis—i.e. motile sperm seen in a test preparation or several sperm detected in the previous histological examination—while enzymatic processing may be the first choice in difficult cases with a high chance of mainly immotile sperm or elongated spermatids.

Interestingly, 188 transfers of embryos derived solely from the injection of immotile testicular sperm resulted in 11 normal pregnancies (Figure 2Go) which represents a much lower pregnancy rate than that obtained with motile testicular sperm (136/522 transfers, 26.1%). Nevertheless this is a considerably better result than the success rates reported in the literature for ejaculated immotile sperm (Nijs et al., 2000Go).

Conclusion
In this retrospective multicentre study no unequivocal advantage of one over the other preparation method could be identified in 839 ICSI cycles using testicular sperm from 549 patients. Fertilization rates with immotile sperm and elongated spermatids seemed to be better after enzymatic versus mechanical treatment and overall embryo quality, judged by the degree of fragmentation, also appeared to be superior after collagenase digestion. Nevertheless, fertilization rates with motile sperm from fresh testicular biopsies were higher after mechanical preparation. Most importantly, resulting pregnancy rates were practically identical. More significant than the method of preparation seems to be the availability of motile testicular sperm for injection to ensure a comparable pregnancy chance as achievable with ejaculated sperm.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The author wishes to thank all contributors to this study for their time and effort in putting all the data together. Special thanks go to my colleagues M.Montag (University of Bonn) and M.Greuner (GMP Happel, Thaele, Happel, Saarbrücken) for careful revision of the manuscript and helpful comments.


    Notes
 
To whom correspondence should be addressed: Vera Baukloh, Fertility Center Hamburg, Speersort 4, D-20095 Hamburg, Germany. E-mail: vbaukloh{at}fertility-center-hh.de Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Crabbe, E., Verheyen, G., Tournaye, H. and Van Steirteghem, A. (1997) The use of enzymatic procedures to recover testicular germ cells. Hum. Reprod., 12, 1682–1687.[Abstract]

Devroey, P., Liu, J., Nagy, Z., Goossens, A., Tournaye, H., Camus, M., Van Steirteghem, A. and Silber, S. (1995) Pregnancies after testicular sperm extraction and intracytoplasmic sperm injection in non-obstructive azoospermia. Hum. Reprod., 10, 1457–1460.[Abstract]

Deutsches IVF Register, (1997), Jahrbuch 1997. D.I.R. Bundesgeschäftsstelle Ärztekammer, Schleswig-Holzstein, Bad Segeberg, p. 13.

Fischer, R., Baukloh, V., Naether, O.G.J., Schulze, W., Salzbrunn, A. and Benson, D.M. (1996) Pregnancy after intracytoplasmic sperm injection of spermatozoa extracted from frozen–thawed testicular biopsy. Hum. Reprod., 11, 2197–2199.[Abstract]

Friedler, S., Raziel, A., Soffer, Y., Strassburger, D., Komarovsky, D. and Ron-El, R. (1997) Intracytoplasmic injection of fresh and cryopreserved testicular spermatozoa in patients with nonobstructive azoospermia – a comparative study. Fertil. Steril., 68, 892–897.[ISI][Medline]

Habermann, H., Seo, R., Cieslak, J., Niederberger, C., Prins, G.S. and Ross, L. (2000) In vitro fertilization outcomes after intracytoplasmic sperm injection with fresh or frozen–thawed testicular spermatozoa. Fertil. Steril., 73, 955–960.[ISI][Medline]

Huang, F.J., Chang, S.Y., Tsai, M.Y., Kung, F.T., Lin, Y.C., Wu, J.F. and Lu, Y.J. (2000) Clinical implications of intracytoplasmic sperm injection using cryopreserved testicular spermatozoa from men with azoospermia. J. Reprod. Med., 45, 310–316.[ISI][Medline]

Küpker, W., Schlegel, P.N., Al-Hasani, S., Fornara, P., Johannisson, R., Sandmann, J., Schill, T., Bals-Pratsch, M., Ludwig, M. and Diedrich, K. (2000) Use of frozen–thawed testicular sperm for intracytoplasmic sperm injection. Fertil. Steril., 73, 453–458.[ISI][Medline]

Nagy, Z., Liu, J., Janssenwillen, C., Silber, S., Devroey, P. and Van Steirteghem, A. (1995) Using ejaculated, fresh, and frozen–thawed epididymal and testicular spermatozoa gives rise to comparable results after intracytoplasmic sperm injection. Fertil. Steril., 63, 808–815.[ISI][Medline]

Nijs, M.H. and Van Der Elst, C.J. (2000) Biological aspects of testicular sperm extraction. Eur. J. Obstet. Gynecol. Reprod. Biol., 92, 1–6.[ISI]

Nijs, M., Vanderzwalmen, P., Vandamme, B., Segal-Bertin, G., Lejeune, B., Segal, L., van Roosendaal, E. and Schoysman, R. (2000) Fertilizing ability of immotile spermatozoa after intracytoplasmic sperm injection. Hum. Reprod., 11, 2180–2185.[Abstract]

Palermo, G., Joris, H., Devroey, P. and Van Steirteghem, A. (1992) Pregnancies after intracytoplasmic injection of single spermatozoa into an oocyte. Lancet, 340, 17–18.[ISI][Medline]

Palermo, G.D., Schlegel, P.N., Hariprashad, J.J., Ergün, B., Mielnik, A., Zaninovic, N., Veeck, L.L. and Rosenwaks, Z. (1999) Fertilization and pregnancy outcome with intracytoplasmic sperm injection for azoospermic men. Hum. Reprod., 14, 741–748.[Abstract/Free Full Text]

Salzbrunn, A., Benson, D.M., Holstein, A.F. and Schulze, W. (1996) A new concept for the extraction of testicular spermatozoa as a tool for assisted fertilization (ICSI). Hum. Reprod., 11, 752–755.[Abstract]

Schulze, W., Knuth, U.A., Jezek, D., Benson, D.M., Fischer, R., Naether, O.G.J., Baukloh, V. and Ivell, R. (1997) Intratesticular sperm extraction. Basis for successful treatment of infertility in men with ejaculatory azoospermia. In Ivell, R. and Holstein, A.F. (eds) The Fate of the Male Germ Cell. Plenum Press, New York, USA, pp. 81–87.

Silber S.J., Nagy, Z., Devroey, P., Camus, M. and Van Steirteghem, A.C. (1997) The effect of female age and ovarian reserve on pregnancy rate in male infertility: treatment of azoospermia with sperm retrieval and intracytoplasmic sperm injection. Hum. Reprod., 12, 2693–2700.[Abstract]

Van Steirteghem, A., Nagy, Z., Joris, H., Janssenwillen, C., Staessen C., Verheyen, G., Camus, M., Tournaye, H., and Devroey, P. (1998) Results of intracytoplasmic sperm injection with ejaculated, fresh and frozen–thawed epididymal and testicular spermatozoa. Hum. Reprod., 13, 134–142.[ISI][Medline]

Verheyen, G., De Croo, I., Tournaye, H., Pletinex, I., Devroye, P. and Van Steirteghem, A.C. (1995) Comparison of four mechanical methods to retrieve spermatozoa from testicular tissue. Hum. Reprod., 10, 2956–2959.[Abstract]

Submitted on February 20, 2001; resubmitted on December 31, 2001; accepted on March 7, 2002.