May–Grünwald–Giemsa stain for detection of spermatogenic cells in the ejaculate: a simple predictive parameter for successful testicular sperm retrieval

Medhat Amer1,2,4, Taha Abd Elnasser1,2, Shawky El Haggar1, Taymour Mostafa1, Geirgis Abdel-Malak2,3 and Wael Zohdy1,2

1 Andrology Department, Cairo University Hospital, Cairo, 2 Adam International Clinic, Guiza and 3 AI/ET Department, Animal Reproduction Research Institute, Guiza, Egypt


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Testicular sperm extraction (TESE) and intracytoplasmic sperm injection (ICSI) have become standard treatments for patients with non-obstructive azoospermia. A diagnostic testicular biopsy for histopathological examination is not always predictive of TESE outcome. Moreover, it is not without potential complications. The aim of this study was to determine the value of various clinical and laboratory parameters, particularly identification of seminal spermatids using May–Grünwald–Giemsa (MGG) stain in predicting TESE results. METHODS: A total of 100 patients with non-obstructive azoospermia was subjected to clinical examination, serum FSH measurement, identification of seminal spermatids and spermatocytes using MGG staining and TESE with multiple testicular sampling. Spermatozoa were retrieved from 49% of patients. Results of TESE were compared with previous parameters in addition to histopathology. RESULTS: Testicular histopathology was, in general, an inaccurate parameter, and identification of testicular spermatids by histology predicted successful TESE in only 74% of cases. Testicular volume and serum FSH concentration also had poor predictive values. Round spermatids were identified in the ejaculate of 83.7% of TESE-positive cases, and in 22% of TESE-negative cases. CONCLUSIONS: The detection of round spermatids in semen by MGG staining provides the greatest predictive value for successful testicular sperm retrieval, and also has the advantages of simplicity, low cost and availability.

Key words: azoospermia/intracytoplasmic sperm injection/May-Grünwald-Giemsa stain/spermatids/sperm extraction


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
It is well established that mature testicular spermatozoa may be found in cases of non-obstructive azoospermia (NOA) (Tournaye et al., 1996Go). Thus, testicular sperm extraction (TESE) combined with intracytoplasmic sperm injection (ICSI) offers azoospermic patients the possibility of fathering their own genetic children, even in the absence of normal spermatogenesis (Devroey et al., 1995Go). TESE may not always be successful in such patients however, as they have minute foci of active spermatogenesis from which a very small number of spermatozoa can be extracted (Silber et al., 1995Go). In addition, multiple biopsies may be necessary in these patients, as this might enhance the probability of sperm retrieval in absolute testicular failure and increase the number of retrieved spermatozoa (Tournaye et al., 1996Go; Hauser et al., 1998Go; Ostad et al., 1998Go; Amer et al., 1999Go). Testicular biopsy is not a completely safe procedure, as it may be followed by inflammatory changes, haematoma, parenchymal fibrosis and even permanent testicular devascularization (Harrington et al., 1996Go; Schlegel and Su, 1997Go; Amer et al., 2000Go). Moreover, repetition of TESE in NOA patients may be difficult, and is even less successful than the first trial (Amer et al., 1999Go).

In order to predict the likelihood of successful TESE, several studies have been performed to determine the value of various parameters, including testis size (Tournaye et al., 1996Go), FSH concentration (Kahraman et al., 1996Go), previous testicular histopathology (Tournaye et al., 1997Go) and, more recently, seminal anti-Müllerian hormone (AMH) (Fenichel et al., 1999Go). There was no strong predictor for successful TESE except testicular histopathology. Recently, it has been reported that spermatozoa can be found in all patients, with detectable round spermatids in testis biopsy (Silber and Johnson, 1998Go). Despite a strong association of testicular spermatids with TESE confirmed, and a prospective study correlating TESE with morphological, biophysical and endocrine profiles in men with functional azoospermia, the correlation between histological patterns and TESE was found to be poor (Ezeh et al., 1998Go). Nonetheless, a good correlation was found between the TESE result and the detection of seminal fluid spermatids, using immunofluorescent localization. However, this technique is sophisticated and requires special skills; hence its routine laboratory application is not straightforward.

Consequently, prediction of TESE outcome, based on seminal characteristics, would be very beneficial in patients with NOA as it may circumvent the need for any preliminary diagnostic testicular biopsy, with its recognized hazards. The presence of ejaculated spermatids indicates that spermatogenesis must have reached the post-meiotic stage in at least part of the testis, and so the possibility of finding testicular spermatozoa without performing diagnostic biopsy is increased. The aim of the current study was to identify seminal spermatids by using the inexpensive, commercially available May– Grünwald–Giemsa (MGG) stain, and to assess its predictive value for successful TESE in comparison with serum FSH concentration, testicular volume and histopathology.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 100 patients with primary infertility was diagnosed as having NOA based on clinical data of testicular size, serum FSH concentration and/or previous histopathology. Seminal biochemical markers were monitored and/or karyotyping was performed in some patients in whom clinical data were inconclusive. All patients provided a full history and details of their current situation. A full clinical examination was carried out at which testicular size was measured using an orchidometer (Seager, Switzerland). On the basis of testis size, patients were classified into three groups: normal-sized testes (>=15 ml); moderate-sized testes (10–15 ml); and small-sized testes (<10 ml). Serum FSH concentration was measured using a radioimmunoassay in those cases where FSH was not a selection criterion. Two consecutive semen analyses were performed to confirm the presence of azoospermia; seminal fluid cells were sedimented (600 g for 15 min), and the pellet was stained with MGG stain (Ludwig and Frick, 1987Go) to detect the presence of spermatogenic cells and identify their types. This was not possible in one aspermic patient who failed to produce semen after electroejaculation. A testicular sperm extraction trial, on occasion with multiple biopsies from each testis (maximum four per testis) was carried out. If no spermatozoa could be found in the testicular biopsies, the biopsy material was centrifuged (400 g for 5 min) and the pellet stained with MGG stain.

Technique of MGG staining
The seminal fluid or testicular biopsy material was centrifuged, and the pellet smeared onto a glass slide and left to air-dry for 1–2 h. The smear was then stained using a stepwise procedure (Ludwig and Frick, 1987Go). The smear was fixed for 15 min in methanol, and transferred without blotting to diluted May–Grünwald solution for 15 min. (A stock 0.3% May–Grünwald solution was prepared by grinding the powdered stain in methanol with a pestle and mortar, and filtering after 2–3 days. For use, the stock solution was diluted 1:1 with phosphate buffer, pH 7.2, the dilute solution being discarded after 1 day.)

Excess stain was drained off on filter paper (without blotting), and the smear was then transferred to dilute Giemsa solution for 30 min. (Giemsa was prepared by dissolving 0.6 g powder in 50 ml methanol, adding 25 ml glycerine, and filtering after 2–3 days. For use, the stock solution was diluted 1:9 in phosphate buffer, pH 7.2.)

The smear was transferred to phosphate buffer, pH 7.2, rinsed for 10–20 s, and then dried and mounted. Smears were examined under an oil immersion lens (x1000) for the presence of spermatocytes and round spermatids according to morphological criteria and staining properties. Primary spermatocytes are broad cells, and have a large central nucleus 8–9 µm in diameter in which the chromosomal spindle and a large central nucleolus 1–2 µm in diameter can be recognized. They are very fragile cells, and are easily damaged by preparation techniques. Initial (round) spermatids are small round cells with a condensed nucleus (sometimes two nuclei) with a continuous rim of cytoplasm and an acrosomal vesicle appearing either as an empty dot or having a pink coloration. At earlier stages, the acrosomal vesicle may not be present; thus, its presence confirms the identification of spermatid (Gandini et al., 1999Go) (Figures 1 and 2GoGo).



View larger version (81K):
[in this window]
[in a new window]
 
Figure 1. May–Grünwald–Giemsa-stained smear of semen. A cluster of spermatogenic cells; the arrow indicates a primary spermatocyte. Scale bar = 10 µm.

 


View larger version (98K):
[in this window]
[in a new window]
 
Figure 2. May–Grünwald–Giemsa-stained smear of semen. A cluster of spermatogenic cells; the arrow indicates a round spermatid. Scale bar = 10 µm.

 
MGG staining was carried out on semen in 99 cases (omitting the patient with aspermia, who failed to provide a semen sample after electroejaculation), and on testicular biopsy material in 51 patients with negative TESE.

Testis biopsy
Under general anaesthesia, a vertical incision was made in the median scrotal raphe. One testis was delivered outside the wound, and inspected together with the epididymis. A biopsy was taken from the upper pole of the testis and sent for immediate sperm detection. If no spermatozoa were identified after thorough examination of the entire specimen, a second biopsy was taken from the lower pole. If still no spermatozoa were found, biopsies were taken from the medial and then lateral aspects of the testis. Two further biopsies were taken from the upper and lower poles and fixed in Bouin's fixative for histopathological examination. After adequate haemostasis, the tunica were closed, and the other testis was delivered through the same incision, whereupon the same steps were repeated. If at any time spermatozoa were seen, the procedure was terminated and the wound closed in layers with 3/0 Vicryl sutures.

TESE
TESE was performed by mincing and shredding of the biopsy specimen (Verheyen et al., 1995Go) in HEPES-buffered Earle's balanced salt solution, using sterile glass slides and forceps under laminar flow in a Petri dish. The minced tissue was examined under an inverted microscope (Hoffman optics) using x400 magnification. If no spermatozoa were seen after checking the entire dish, the contents were transferred to a 5 ml Falcon tube after removing gross testicular debris, centrifuged at 400 g for 5 min, and the pellet was re-examined for the presence of spermatozoa. The result was interpreted as positive or negative TESE according to the presence or absence of at least one live spermatozoon; this was either spontaneously motile, became motile after pentoxifylline stimulation, or was shown to be viable by the hypo-osmotic swelling test (HOST).

Statistical analysis
{chi}2 and unpaired Student's t-tests were used as appropriate. Statistical significance was accepted at a P value of < 0.05.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The mean age of patients was 36.43 ± 5.83 (range 23–50) years, and the mean serum FSH concentration was 16. 26 ± 11.83 (range 2.7–56.0) IU/l (normal range = 2.0–8.0 IU/l). The mean testicular volume was 12.70 ± 3.65 ml on the right side, and 12.73 ± 3.49 ml on the left side. Biopsies were taken from both sides in 69 patients, from only the right side in 17 patients, and from only the left side in 14 patients. Histopathology of the patients was classified into five groups: early spermatid arrest; primary spermatocyte (C1) arrest; mixed pathology; complete Sertoli cell-only (SCO) syndrome; and Klinefelter (KF) -like (Table IGo). Differences in histopathology were found between the right and left sides in 15 cases (21.74%) biopsied bilaterally, while differences between upper and lower poles were detected only in 6.42% of biopsied testes.


View this table:
[in this window]
[in a new window]
 
Table I. Incidence of histopathological findings in 100 patients with functional azoospermia, and TESE outcome
 
Spermatozoa were successfully retrieved in 49 patients; from the right side in 22 patients, from the left side in 26, and from both sides in one patient where the other testis was biopsied while searching for motile spermatozoa. The results of TESE according to histopathology are shown in Table IGo.

There was no statistically significant difference in testicular volume between patients with retrievable spermatozoa and those without (Table IIGo). Furthermore, no lower limit for testicular volume to exclude the presence of spermatozoa was identified, where spermatozoa could be retrieved from testes of <5 ml volume.


View this table:
[in this window]
[in a new window]
 
Table II. Comparison between testicular volume and TESE findings
 
On the basis of FSH concentration, patients were classified into three groups (Table IIIGo). There was no statistically significant difference in TESE outcome between groups I (normal FSH) and II (mildly elevated FSH), but a significant difference was found between groups I and III (markedly elevated FSH) (P = 0.013). A statistically significant difference was also found in mean FSH concentration between TESE-positive and TESE-negative cases (P = 0.027). The sensitivity of FSH concentration as a laboratory parameter to predict TESE outcome was low (36.7%) when patients were classified as normal and above-normal concentrations (Table IVGo). However, when higher FSH concentration was considered, the sensitivity increased while the specificity decreased (Table VGo).


View this table:
[in this window]
[in a new window]
 
Table III. TESE outcome in relation to serum FSH concentration
 

View this table:
[in this window]
[in a new window]
 
Table IV. Sensitivitya and specificityb of serum FSH concentration in predicting TESE outcome (threshold value 8 IU/l)
 

View this table:
[in this window]
[in a new window]
 
Table V. Sensitivitya and specificityb of serum FSH concentration in predicting TESE outcome (threshold value IU/l)
 
With MGG staining, primary spermatocytes were detected in the ejaculate of 74 patients. Primary spermatocytes were also found in semen of all patients with primary spermatocyte and early spermatid arrest and patients with mixed pathology, while in patients with SCO and KF-like histopathology, they were detected in only 48.3 and 37.5% of patients respectively. Round spermatids were detected in the ejaculate of 52 patients, in 80% of cases with early spermatid arrest, 44.4% of C1 arrest, 88% of mixed pathology, 20.7% of SCO syndrome and in 25% of cases with KF-like histopathology. With regard to TESE outcome, primary spermatocytes were detected in the ejaculate of all TESE-positive cases, and in the ejaculate of 50% of TESE-negative cases (Table VIGo). Round spermatids were detected in 83.7% of TESE-positive cases, and in 22% of TESE-negative cases (Table VIIGo).


View this table:
[in this window]
[in a new window]
 
Table VI. Sensitivitya and specificityb of primary spermatocyte (C1) detection in semen in predicting TESE outcome
 

View this table:
[in this window]
[in a new window]
 
Table VII. Sensitivitya and specificityb of round spermatid detection in semen in predicting TESE outcome
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
ICSI of the oocyte using a testicular spermatozoon has enabled the treatment of patients with azoospermia due to primary gonadal failure (Tournaye et al., 1996Go). Because testicular biopsy may be associated with significant potential complications (Harrington et al., 1996Go; Schlegel and Su, 1997Go), attempts have been made to render this procedure more selective. However, the prediction of successful sperm retrieval with current laboratory and clinical methods has been disappointing.

Testicular volume was found to have a poor predictive value for successful TESE, as topographical variations in testicular pathology can occur independently of testicular size. Thus, large testicular volume does not indicate successful TESE in patients showing SCO pattern in their testicular histopathology. Conversely, in patients showing maturation arrest pattern, a low testicular volume does not preclude successful TESE (Devroey et al., 1995Go). On the basis of the current results, testicular volume is seen to be an inaccurate parameter, as spermatozoa were successfully retrieved from 44% of normal-sized testes, 42% of moderate-sized testes, and from 25% of small-sized testes, the latter value being especially high for such a population of patients. Furthermore, no lower limit of testicular volume excluded the presence of testicular spermatozoa.

In the current study it was possible to retrieve spermatozoa from 62% of patients with normal FSH, from 59.4% with mildly elevated FSH, and from 30.8% with markedly elevated serum FSH. No upper limit for FSH could be detected that excluded the presence of testicular spermatozoa. The sensitivity of FSH in predicting successful TESE is low at normal levels, but with higher concentrations of FSH the sensitivity increases while the specificity decreases. These results support the findings of many investigators who showed that serum FSH concentration has a poor predictive value for successful TESE (Chen et al., 1996Go; Kahraman et al., 1996Go; Jezek et al., 1998Go; Ezeh et al., 1999Go). This is because FSH concentration is not related to sperm count or production, but it is related inversely to the total number of testicular germ cells present. Hence, the only relationship between serum FSH concentration and the more advanced stages of spermatogenesis is an indirect one, and depends strictly on the primary inverse relationship between FSH and spermatogonia (Silber et al., 1996Go). Therefore, patients with complete maturation arrest could be expected to have normal FSH concentrations, but no spermatozoa could be retrieved from the testis, whereas patients with SCO pattern in most of the testis and with small foci of active spermatogenesis are most likely to have a very high FSH concentration, with spermatozoa retrievable on biopsy (Silber et al., 1996Go).

Histopathology of testicular biopsy sections is considered the best predictive parameter for successful TESE, especially with visualization of mature testicular spermatids. Although its value in functional azoospermia is questionable (Tournaye et al., 1996Go, 1997Go), visualization of testicular spermatids at histopathological examination was found to provide a correct prediction in 77% of cases (Ezeh et al., 1999Go).

Correlation of TESE with the histopathological picture showed that spermatozoa were retrievable from 85% of cases with round spermatid arrest, from 44.4% in C1 arrest, from 65% with mixed pathology, from 24.2% in SCO syndrome and from 25% in cases with tubular sclerosis. Thus, the chances of sperm retrieval improves with more advanced stages of spermatogenesis, and detection of round spermatids in histopathology can predict successful TESE in about 74% of the current cases. However, spermatozoa could still be retrieved from cases without any spermatids seen histopathologically. Consequently, histopathology alone cannot be considered as an accurate parameter for prediction when spermatids are not observed, even in cases of round spermatid arrest, when spermatozoa could not be detected in 15% of cases. This fact confirms our previous description of the entity of complete spermiogenesis failure, where round spermatids were detected in the testis without finding testicular spermatozoa (Amer et al., 1997Go). The finding also confirms that testicular biopsy for histopathology is not completely representative of the whole testis, there being variation between both sides, and also from site to site within the same testis (Amer et al., 1999Go).

Immature germ cells may be present in the ejaculate of normospermic, oligozoospermic and azoospermic patients. The quantity of these germ cells increases as the sperm count decreases, and is not related to serum FSH concentration (Tomlinson et al., 1992Go; Tesarik and Mendoza, 1996Go). It has also been reported that the presence of round spermatids in testis biopsy indicates the presence of mature spermatozoa, and it was claimed that no case with round spermatids has complete absence of spermatozoa (Silber and Johnson, 1998Go). Furthermore, the number of testicular spermatids has been shown to correlate with sperm count in men with oligozoospermia (Silber and Rodriguez, 1981Go), and with testicular spermatozoa in men with non-obstructive azoospermia (Mulhall et al., 1997Go; Silber et al., 1997Go). In a recent study (Ezeh et al., 1999Go), it was reported that identification of seminal spermatids by an immunofluorescence method utilizing mouse monoclonal antibodies could predict testicular sperm retrieval (100%); however, testicular spermatozoa could still be extracted from patients without detectable seminal spermatids in 25% of cases. Thus, the sensitivity was 100% while the specificity was 75%. Various methods ranging from Percoll centrifugation and direct microscopy of cells in the ejaculates (Tesarik et al., 1995Go; Tesarik and Mendoza, 1996Go; Gandini et al., 1999Go) to immunological methods such as staining with monoclonal antibodies to the acrosomal moiety (Moore et al., 1987Go) and antiacrosin antibodies (Tesarik and Mendoza, 1996Go) have been used to identify spermatids in the human ejaculate. These techniques are rather tedious, expensive and require special facilities, however. A more simple technique was described by using May–Grünwald–Giemsa stain to identify various types of testicular and ejaculated germ cells (Ludwig and Frick, 1987Go; Foresta et al., 1992Go; Gandini et al., 1999Go). This method has been found to be simple, effective and inexpensive, and also to correlate with other methods of spermatogenic cell identification such as Papanicolaou (Pap) staining and immunofluorescence techniques (M.Amer et al., unpublished data).

Evidence is accumulating that deletion of the AZFc region only on the Y chromosome confers a slightly better chance of having usable spermatozoa, either from the ejaculate or from the testis, whereas AZFb-region deletions may predict failure of sperm recovery (Brandell et al., 1998Go). The current results with MGG showed that identification of round spermatids in semen has a sensitivity of 83.7% and a specificity of 78%, while identification of primary spermatocytes has a sensitivity of 100% and a specificity of only 50%. Thus, this simple, cheap test can predict successful TESE in comparison with the expensive sophisticated methods used to detect Y-chromosome microdeletion and the azoospermia factor. Moreover, the detection of round spermatids in semen using MGG stain as a predictive parameter for successful sperm recovery is more accurate than testicular volume, FSH concentration or even previous histopathology or tested TESE, as the patient may have spermatozoa in previous tested TESE, but no spermatozoa could be retrieved at the time of TESE–ICSI attempts (Vanderzwalmen et al., 1997Go; Amer et al., 1999Go). More extensive methods for processing testicular biopsy using collagenase and trypsin inhibitor or erythrocyte-lysing buffer provide a gentle dissolution of the cells from their tissue, and clear the red blood corpuscles. This results in high yields of mature viable spermatozoa, and makes the search for spermatozoa simple and more rapid (Salzbrunn et al., 1996Go; Crabbé et al., 1997Go; Nagy et al., 1997Go). Thus, we expect that microsurgical biopsy to select healthy tubules (Schlegel, 1999Go; Amer et al., 2000Go), combined with more extensive processing of the biopsy using the above-mentioned techniques, providing more time and patience for the search, will increase the sensitivity and specificity of spermatid identification in predicting TESE outcome and the numbers of false-negative cases will be minimized.

The fact that motile spermatozoa were recovered from cases without detected spermatids (eight patients) in the ejaculate may suggest the presence of partial or intermittent obstruction. We agree with others (Ezeh et al., 1998Go), that the absence of ejaculated spermatids cannot be used to distinguish obstructive from non-obstructive azoospermia, but that their presence indicate NOA. Another hypothesis which had been advanced (Ezeh et al., 1998Go) is that there may be a threshold level of spermatogenesis below which no spermatids could be detected in semen. In support of this hypothesis is a study showing that in order for any spermatozoa to reach the ejaculate, more than six mature spermatids per seminiferous tubule must be present in the histological section of the testis (Silber et al., 1997Go). Moreover, MGG staining after a modified discontinuous Percoll gradient technique (Gandini et al., 1999Go) may be helpful in detecting spermatids in patients with negative MGG screening test in semen, and retrievable testicular spermatozoa.

The current results indicate that the sensitivity of detecting seminal spermatids is 100% in predicting the presence of testicular round spermatids, and its specificity is 92.8%. This observation may be important in the selection of cases who are candidates for in-vitro culture of spermatogenic cells from testicular tissue in case of complete absence of spermatozoa after an extensive search, as has been recently suggested (Tesarik et al., 1998Go).

In conclusion, the detection of seminal spermatids using MGG stain offers a good predictive parameter for successful TESE compared with testicular size, serum FSH concentration and histopathological diagnostic testicular biopsy. Additional advantages of the MGG stain are the simplicity, reliability, low cost and commercial availability. Advances in the techniques of TESE and processing combined with extensive search may further improve its sensitivity and specificity in the near future.


    Notes
 
4 To whom correspondence should be addressed at: Adam International Clinic, 20 Aden street, P. O. Box 12411, Mohandessin, Guiza, Egypt. E-mail: mkamer{at}mednet3.camed.eun.eg Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Amer, M., Soliman, E., El-Sadek, M. et al. (1997) Is complete spermiogenesis failure a good indication for spermatid conception? Lancet, 350, 116.

Amer, M., El-Haggar, S., Abd El-Nasser, T. et al. (1999) Testicular sperm extraction: impact of testicular histology on outcome, number of biopsies to be performed and optimal time for repetition. Hum. Reprod., 14, 3030–3034.[Abstract/Free Full Text]

Amer, M., Attia, A., Ragab, H. and Zohdy, W. (2000) Prospective comparative study between microsurgical and conventional testicular sperm extraction: follow up by serial ultrasound examination. Hum. Reprod., 15, 653–656.[Abstract/Free Full Text]

Brandell, R.A., Mielnik, A., Liotta, D. et al. (1998) AZFb deletions predict the absence of spermatozoa with testicular sperm extraction: preliminary report of a prognostic genetic test. Hum. Reprod., 13, 2812–2815.[Abstract/Free Full Text]

Chen, C.S., Chu, S.H., Lai, Y.M. et al. (1996) Reconsideration of testicular biopsy and follicle-stimulating hormone measurement in the era of intracytoplasmic sperm injection for non obstructive azoospermia. Hum. Reprod., 11, 2176–2179.[Abstract]

Crabbé, E., Verheyne, G., Tournaye, H. et al. (1997) The use of enzymatic procedures to recover testicular germ cells. Hum. Reprod., 12, 1682–1687.[Abstract]

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.[Abstract]

Ezeh, U.I.O., Martin, M., Cook, I.D. et al. (1998) Correlation of testicular pathology and sperm extraction in azoospermic men with ejaculated spermatids detected by immunofluorescent localisation. Hum. Reprod., 13, 3061–3065.[Abstract]

Ezeh, U.I.O., Taub, N.A., Moore, H.D.M. et al. (1999) Establishment of predictive variables associated with testicular sperm retrieval in men with non obstructive azoospermia. Hum. Reprod., 14, 1005–1012.[Abstract/Free Full Text]

Fenichel, P., Rey, R., Poggioli, S. et al. (1999) Anti-Mullerian hormone as a seminal marker for spermatogenesis in non obstructive azoospermia. Hum. Reprod., 14, 2020–2024.[Abstract/Free Full Text]

Foresta, C., Varotto, A. and Scandellari, C. (1992) Assessment of testicular cytology by fine needle aspiration as a diagnostic parameter in the evaluation of azoospermic subject. Fertil. Steril., 57, 858–865.[ISI][Medline]

Gandini, L., Lenzi, A., Lombardo, F. et al. (1999) Immature germ cell separation using a modified discontinuous Percoll gradient technique in human semen. Hum. Reprod., 14, 1022–1027.[Abstract/Free Full Text]

Harrington, T.G., Schaure, D. and Gilbert, B.R. (1996) Percutaneous testis biopsy: an alternative to open testicular biopsy in the evaluation of subfertile man. J. Urol., 156, 1647–1652.[ISI][Medline]

Hauser, R., Botchon, A., Amit, A. et al. (1998) Multiple testicular sampling in non obstructive azoospermia, is it necessary? Hum. Reprod., 13, 3081–3085.[Abstract]

Jezek, D., Knuth, V.A. and Schulze, W. (1998) Successful testicular sperm extraction (TESE) in spite of high serum follicle stimulating hormone and azoospermia. Correlation between testicular morphology, TESE results, semen analysis and serum hormone values in 103 infertile men. Hum. Reprod., 13, 1230–1235.[Abstract]

Kahraman, S., Ozgur, S. and Alatas, C. (1996) High implantation and pregnancy rates with testicular sperm extraction and intracytoplasmic sperm injection in obstructive and non-obstructive azoospermia. Hum. Reprod., 11, 673–679.[Abstract]

Ludwig, G. and Frick, J. (1987) Ejaculate analysis (spermiogram). In Ludwig, G. and Frick, J. (eds) Spermatology. Springer-Verlag, Berlin, Heidelberg, pp. 30–32.

Moore, H.D.M., Smith, C.A., Hartman, T.D. et al. (1987) Visualization and characterisation of the acrosome reaction to human spermatozoa by immunolocalisation with monoclonal antibody. Gamete Res., 17, 245–259.[ISI][Medline]

Mulhall, J.P., Burgess, C.M., Cunninghan, D. et al. (1997) The presence of mature sperm in testicular parenchyma of men with non obstructive azoospermia: prevalence and predictive factors. Urology, 49, 91–95.[ISI][Medline]

Nagy, Z.P., Devroey, P., Verheyen, G. et al. (1997) An improved treatment procedure for testicular biopsy specimens offers more efficient sperm recovery: case series. Fertil. Steril., 68, 376–379.[ISI][Medline]

Ostad, M., Liotta, D., Ye, Z. et al. (1998) Testicular sperm extraction (TESE) for non-obstructive azoospermia: results of multibiopsy approach with optimized tissue dispersion. Urology, 52, 692–696.[ISI][Medline]

Salzbrunn, A., Benson, D.M., Holstein, A. 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]

Schlegel, P.N. (1999) Testicular sperm extraction: microdissection improves sperm yield with minimal tissue excision. Hum. Reprod., 14, 131–135.[Abstract/Free Full Text]

Schlegel, P. and Su, L. (1997) Physiological consequences of testicular sperm extraction. Hum. Reprod., 12, 1688–1692.[Abstract]

Silber, S.J. and Rodriguez, L.I. (1981) Quantitative analysis of testicle biopsy: determination of partial obstruction and prediction of sperm count after surgery for obstruction. Fertil. Steril., 36, 480–485.[ISI][Medline]

Silber, S.J. and Johnson, L. (1998) Are spermatid injection of any clinical value? ROSNI and ROSI revisited. Hum. Reprod., 13, 509–515.[Free Full Text]

Silber, S.J., Van Steirteghem, A.C. and Devroey, P. (1995) Sertoli cell only revisited. Hum. Reprod., 10, 1031–1036.[ISI][Medline]

Silber, S.J., Van Steirteghem, A.C., Nagy, Z. et al. (1996) Normal pregnancies resulting from testicular sperm extraction and intracytoplasmic sperm injection for azoospermia due to maturation arrest. Fertil. Steril., 66, 110–117.[ISI][Medline]

Silber, S.J., Nagy, Z., Devroey, P. et al. (1997) Distribution of spermatogenesis in the testicles of azoospermic men: the presence or absence of spermatids in the testes of men with germinal failure. Hum. Reprod., 12, 2422–2428.[Abstract]

Tesarik, J. and Mendoza, C. (1996) Spermatid injection into human oocytes. I. Laboratory techniques and special features of zygote development. Hum. Reprod., 11, 772–779.[Abstract]

Tesarik, J., Mendoza, C. and Testart, J. (1995) Viable embryos from injection of round spermatids into oocytes [letter]. N. Engl. J. Med., 333, 525.[Free Full Text]

Tesarik, J., Ermanno, G., Rienzi, L. et al. (1998) Differentiation of spermatogenic cells during in-vitro culture of testicular biopsy samples from patients with obstructive azoospermia: effect of recombinant follicle stimulating hormone. Hum. Reprod., 13, 2772–2781.[Abstract/Free Full Text]

Tomlinson, M.I., Barratt, C.L.R. and Bolton, A.E. (1992) Round cells and sperm fertilizing capacity: the presence of immature germ cells but not seminal leukocytes are associated with reduced success of in-vitro fertilization. Fertil. Steril., 58, 1257–1264.[ISI][Medline]

Tournaye, H., Liu, J., Nagy, Z. et al. (1996) Correlation between testicular histology and outcome after intracytoplasmic sperm injection using testicular sperm. Hum. Reprod., 11, 123–127.

Tournaye, H., Verheyen, G., Nagy, Z. et al. (1997) Are there any predictive factors for successful testicular sperm recovery in azoospermic patients. Hum. Reprod., 12, 80–86.[ISI][Medline]

Vanderzwalmen, P., Zech, H. and Birkenfeld, A. et al. (1997) Intracytoplasmic injection of spermatids retrieved from testicular tissue: influence of testicular pathology, type of selected spermatids and oocyte activation. Hum. Reprod., 12, 1203–1213.[ISI][Medline]

Verheyen, G., Croo, I., Tournaye, H. et al. (1995) Comparison of four mechanical methods to retrieve spermatozoa from testicular tissue. Hum. Reprod., 10, 2956–2959.[Abstract]

Submitted on June 12, 2000; accepted on March 20, 2001.