H. OBARA,* H. SHIBAHARA,* H. TSUNODA, A. TANEICHI,* H. FUJIWARA,* S. TAKAMIZAWA,* S. IDEI* and I. SATO*

Transcription

1 international journal of andrology, 24:102±108 (2001) Prediction of unexpectedly poor fertilization and pregnancy outcome using the strict criteria for sperm morphology before and after sperm separation in IVF±ET H. OBARA,* H. SHIBAHARA,* H. TSUNODA, A. TANEICHI,* H. FUJIWARA,* S. TAKAMIZAWA,* S. IDEI* and I. SATO* * Department of Obstetrics and Gynaecology, and Department of Clinical Laboratory, Jichi Medical School, Tochigi, Japan Summary This study was performed to investigate if unexpectedly poor fertilization and in-vitro fertilization (IVF) outcome could be predicted using sperm morphology as diagnosed by the strict criteria. Sperm morphology was assessed in 137 IVF±ET cycles with at least three oocytes collected. The lowest amount of normal forms was 5% in 137 samples, indicating there were no patients belonging to `poor prognosis' (<5% normal forms). Treatment using intracytoplasmic sperm injection (ICSI) was also excluded. Before sperm separation by the swim-up method, sperm morphology demonstrated a signi cant correlation with the fertilization rate (p < ). The fertilization rate (80.5%) in 110 `normal' samples (>14% normal forms) was signi cantly higher (p < 0.01) than that (55.4%) in 27 samples with `good prognosis' (those with 5±14% normal forms). No embryo was available for transfer (ET) in 4 (3.6%) of 110 `normal' cycles and in 3 (11.1%) of 27 `good prognosis' cycles (not signi cant). Fresh ET was intentionally cancelled to avoid severe ovarian hyperstimulation syndrome (OHSS) in six of 110 `normal' cycles as well as in one of 27 `good prognosis' cycles. The pregnancy rate per ET was 31.0% (31/ 100) in the former group, while it was 26.1% (6/23) in the latter group. There was no difference between the two groups. In the post swim-up evaluation of sperm characteristics, morphology was signi cantly correlated with the fertilization rate in IVF± ET (p < 0.05) while other sperm parameters were not. When the cut off level for the post swim-up sperm morphology was set at 25%, there was a signi cant difference in the fertilization rates between patients (78.6%) with post-swim-up >25% and those (55.0%) with post-swim-up 25% (p < 0.01). Taken together, a relative indication for ICSI using sperm morphology before and after swim-up was established. Category A includes 14% normal forms in the ejaculate and post-swim-up 25%, while Category B includes 14% in the ejaculate and post-swim-up >25%. There was a signi cant difference in the fertlization rates between patients (47.2%) in Category A and those (60.2%) in Category B (p < 0.05). The clinical pregnancy rate was 11.1% for patients in Category A compared with 35.7% for patients in Category B. However, there was no signi cant difference between the two categories. These results indicate that the strict criteria provide Correspondence: Assoc. Prof. Hiroaki Shibahara, Department of Obstetrics and Gynaecology, Jichi Medical School, , Yakushiji, Minamikawachi-machi, Kawachi-gun, Tochigi , Japan. sibahara@jichi.ac.jp Ó 2001 Blackwell Science Ltd.

2 Sperm morphology and fertilization 103 a reliable estimation of the fertilizing ability of human spermatozoa. ICSI might be considered in Category A patients to avoid poor fertilization and pregnancy outcome. Keywords: fertilization, IVF±ET, male infertility, spermatozoa, strict morphology Introduction Intracytoplasmic sperm injection (ICSI) has been widely used to overcome severe male infertility. Similar pregnancy rates can be obtained with this new technique as with conventional in-vitro fertilization (IVF) (Van Steirteghem et al., 1993). Although available data on children born after ICSI is encouraging (Bonduelle et al., 1996), the procedure must still be considered as novel and the safety aspects are to a considerable extent unexplored. Hamberger et al. (1998) suggested that absolute indications for ICSI included two previous fertilization failures with conventional IVF, use of epididymal or testicular sperm samples, or when only acrosome-de cient or immotile spermatozoa were available. 1 We previously reported (Shibahara et al., 1998) that the results of the zona-free hamster egg penetration test (HEPT) was well correlated with the fertilizing ability of human spermatozoa in patients treated by conventional IVF, and that couples suffering from male factor infertility with a sperm penetration index of <15 should consider treatment with ICSI. However, it is well-known that the semen characteristics as well as the fertilizing ability of human spermatozoa are not always reproducible. Examination of the functional capacity of human spermatozoa in vitro is probably of greater value in predicting fertility. IVF provides the best means of investigating sperm± egg interaction and estimating fertilization probability for diagnostic purposes. However, it cannot obviously be used as a routine screening test. As the predictive value of conventional semen parameters appears to be limited, several discriminatory tests that can foretell the fertilizing potential of spermatozoa have been identi ed. The hamster oocyte penetration test (Yanagimachi et al., 1976), the hemi-zona assay (Burkman et al., 1988) and the computer-aided sperm analysis (CASA) are examples which have been used to assess the fertilizing ability of human spermatozoa. Although there are some reports in which sperm morphology did not correlate with fertililization rate in IVF (Alper et al., 1985; Hirsch et al., 1986), others arrived at 2 the opposite conclusion (Mahadevan & Trounson, 1984; 3 Yovich & Stanger, 1984; Jeulin et al., 1986; Chan et al., 1989; Liu et al., 1989). To achieve a better predictive value of fertilization, strict criteria for sperm morphology were proposed by Kruger et al. (1986, 1988), which were later established as the universal predictive value of normal sperm morphology in a structured literature review by Coetzee et al. (1998). The evaluation of sperm morphology using strict criteria is a simple, cost-effective method. By evaluating sperm morphology with the strict criteria, normal morphology becomes a valuable tool via which to predict a patient's chance of fertilization and reaching the transfer stage. In a large IVF population study, Grow et al. (1994) reported that strict morphology was an excellent biomarker of sperm fertilizing capacity, independent of motility and concentration, and that <4% normal forms might denote a poorer prognosis for establishing a pregnancy. Recent studies have found that morphology assessments using native and postseparation spermatozoa were the best predictors of IVF outcome (Enginsu et al., 1993). However, no investigators were able to nd out how to avoid cancellation of embryos transferred (ET) after IVF using a sperm morphology assessment. This study was performed to investigate the predictive value of sperm morphology as diagnosed by the strict criteria before and after sperm separation in IVF outcome. Materials and methods Semen samples One hundred and thirty-seven fresh semen samples from 99 men were obtained by masturbation with at least three oocytes collected in IVF±ET between May 1995 and December All the IVF cycles were performed according to routine IVF procedures as we have previously published (Shibahara et al., 1997, 1998, 1999). A total of 99 women with a mean age of 33.5 years were treated in 137 cycles. The medical indications for conventional IVF treatment were as follows: tubal factors in 46 cases; the wife's sera positive for an immunological factor, such as sperm-immobilizing antibodies, in two cases; unexplained in 42 cases, and a male factor in nine cases. In this study, treatment using ICSI was excluded, indicating that patients with severe male infertility were not included. Routine semen analysis After liquefaction, semen quality analysis was performed using a CASA system (Hamilton Thorne; Hamilton Thorne Research, Beverly MA, USA). Sperm morphology was assessed following the method of Kruger et al. (1988). Brie y, 5 ll semen were used to make the smears as thin as possible using two morphology slides cleaned thoroughly with 70% ethanol before use. The slides were air-dried at room temperature and xed for 15 sec with Diff-Quik xative (Diff-Quik AHS del Caribe, Inc. Aguada, PR

3 104 H. Obara et al ) prior to staining with Diff-Quik solutions. The slides were read and documented on the same day. Spermatozoa were considered normal when the head had a smooth oval con guration with a well-de ned acrosome involving about 40±70% of the sperm head, as well as an absence of neck, midpiece, or tail defects. No cytoplasmic droplets of more than half the size of the sperm head should be present. The length of a normal sperm head was 5±6 lm and the diameter 2.5±3.5 lm. At least 200 cells per slide were evaluated. Patients were divided into three categories on the basis of their morphology according to Kruger et al. (1988): `normal' (>14% normal forms), `good prognosis' (those with 5±14% normal forms), and `poor prognosis' (those with <5% normal forms). Swim-up method In the swim-up procedure, semen samples were mixed with 7 ml Sydney IVF Sperm Buffer (COOK IVF, Australia) containing human serum albumin (HSA) and centrifuged at 569 g (1800 r.p.m.) for 5 min. The pellet was resuspended in 0.5 ml of the same medium. Swim-up from semen was performed using a combined migration± sedimentation method (Lucena et al., 1989) in a BIO±LABO tube (Jyuji Field Co., Tokyo, Japan). Brie y, spermatozoa migrated from semen contained in a ring-shaped well that was completely overlayed with a layer of culture medium. The central hole of the ring constituted the collection well into which motile spermatozoa settled within 1±2 h at 37 C. The upper two-thirds of the supernatant was collected and mixed with Sydney IVF Sperm Medium, followed by centrifugation at 569 g for 5 min. The pellet was resuspended in 0.5 ml of the same medium and the sperm characteristics and morphology were assessed as described above. ET were established, giving a pregnancy rate of 30.1%. The implantation rate was 11.6% (48/415). Semen characteristics of infertile subjects treated with IVF±ET The semen characteristics and sperm morphology for a total of 137 fresh semen samples from 99 men were evaluated. Table 1 shows the semen characteristics. The mean SD for semen volume, sperm concentration, sperm motility, progressive motility and total motile count were ml, /ml, %, % and , respectively. Normal sperm morphology was %. As the lowest normal forms was 5% in 137 samples, no patient belonged to the category of `poor prognosis'. Table 1. Semen characteristics and sperm morphology in 137 fresh samples from 99 infertile subjects treated with IVF±ET Characteristics Mean Volume (ml) (0.5±8.0) Concentration (10 6 /ml) (2.2±521.6) Motility (%) (12.0±92.0) Progressive motility (%) (0.0±79.0) Total motile count (10 6 ) (11.0±1749.9) Normal morphology (%) (5.0±58.0) Values are mean SD, with range in brackets. Statistical analysis Statistical analysis of the data was performed using the v 2 test and Student's t-test (Statview 4.5; Abacus Concepts, Berkeley, CA, USA). Pearson's correlation coef cient was used to analyse the predictability of the fertilization rate by the semen characteristics. Results Overall results in IVF±ET At least three mature oocytes were collected in 137 IVF± ET treatment cycles from 99 infertile women. The average number of oocytes collected was (mean SD). The overall fertilization rate was %. No embryo was available in seven cycles. ET was intentionally cancelled to avoid developing severe ovarian hyperstimulation syndrome (OHSS) in seven cycles. Clinical pregnancies including 29 singles, ve twins and three triplets out of 123 fresh Figure 1. Relationship between the percentage normal sperm morphology and the fertilization rate in IVF±ET.

4 Sperm morphology and fertilization 105 Correlation between the normal sperm morphology and the fertilization rate in IVF±ET The normal sperm morphology and the fertilization rate were compared in the 137 IVF±ET cycles. Sperm morphology demonstrated a signi cant correlation with the fertilization rate (p < ) (Fig. 1). As shown in Table 2, the fertilization rate (80.5%) in 110 `normal' samples was signi cantly higher than that (55.4%) in 27 `good prognosis' samples (p < 0.01). No embryo was available in four (3.6%) of 110 `normal' cycles and in three (11.1%) of 27 `good prognosis' cycles (not signi cant). Fresh ET was intentionally cancelled to avoid severe OHSS in 6 of 110 `normal' cycles as well as in one of 27 `good prognosis' cycles. The pregnancy rate per ET was 31.0% (31/100) in the former group, while it was 26.1% (6/23) in the latter group. There was no signi cant difference between the two groups. Predictability of the IVF fertilization outcome by the semen characteristics and sperm morphology As shown in Table 3, there was no correlation between semen volume and the fertilization rate in IVF±ET. In contrast, sperm concentration, sperm motility, progressive Table 2. The relationship between sperm morphology and fertilization and pregnancy rates Morphology category a Normal Good prognosis Normal forms (%) >14 5±14 <5 No. of IVF cycles Eggs inseminated Eggs fertilized Fertilization rate 80.5 b 55.4 b No. of embroys transferred No. of pregnancy 31 6 Pregnancy rate/et (%) a Adopted from Kruger et al. (1988). b p < Poor prognosis Table 3. Predictability of IVF fertilization outcome according to semen characteristics and sperm morphology Characteristics Correlation Volume (ml) )0.043 Concentration (10 6 /ml) a Motility (%) b Progressive motility (%) c Total motile count (10 6 ) c Normal morphology b motility, total motile count and normal sperm morphology were signi cantly correlated with the fertilization rate in IVF±ET. Post-swim-up sperm characteristics and morphology The post-swim-up sperm characteristics for a total of 137 samples from 99 men were evaluated. As no sperm were available after swim-up in one infertile man, a total of 136 samples from 98 men were examined by sperm morphology. Table 4 shows the semen characteristics. The mean SD for sperm concentration, sperm motility and progressive motility were /ml, % and %, respectively. Normal sperm morphology was %. Predictability of the IVF fertilization outcome by the postswim-up sperm characteristics and morphology As shown in Table 5, there was no correlation between sperm concentration, sperm motility, progressive motility and the fertilization rate in IVF±ET. In contrast, sperm morphology was signi cantly correlated with the fertilization rate in IVF±ET (p < 0.05). Calculation of the best cut-off level for post swim-up sperm morphology When the cut-off level for the post swim-up sperm morphology was set at 25%, there was a signi cant difference in the fertilization rates between patients (78.6%) with postswim up >25% and those (55.0%) with post-swim-up <25% Table 4. Post-swim-up sperm characteristics and morphology in 137 samples from 99 infertile subjects treated with IVF±ET Characteristics n Mean Concentration (10 6 /ml) (0.0±228.8) Motility (%) (0.0±100.0) Progressive motility (%) (0.0±100.0) Normal morphology (%) 136 a (8.0±81.0) Values are mean SD, with range in brackets. a No spermatozoa were available after swim-up in one infertile man. Table 5. Predictability of IVF fertilization outcome according to the post-swim-up sperm characteristics and morphology Characteristics Correlation Concentration (10 6 /ml) Motility (%) Progressive motility (%) Normal morphology (%) a a p < 0.005, b p < 0.001, c p < 0.05 (Pearson's coef cient). a p < 0.05 (Pearson's coef cient).

5 106 H. Obara et al. (p < 0.01) (Table 6). The threshold post swim-up of 25% was chosen because the least p-value (0.0077) was available at this point. Establishment of a relative indication for ICSI using sperm morphology strict criteria before and after swim-up To establish a relative indication for ICSI using the sperm morphology strict criteria before and after swim-up, especially in 27 patients with `good prognosis', they were divided into two categories based on the percentage normal sperm morphology: Category A: ejaculated 14% and post-swim-up 25%, Category B: ejaculated 14% and post-swim-up >25%. As shown in Table 7, there was a signi cant difference (p < 0.05) in the fertilization rates between patients (47.2%) in Category A and those (60.2%) in Category B. The clinical pregnancy rate was 11.1% for patients in Category A and 35.7% in Category B. However, there was no signi cant difference between the two categories. Discussion The ICSI was introduced clinically with the aim to overcome severe male infertility, giving fertilization and term pregnancy rates similar to conventional IVF. However, it should be emphasized that ICSI is used only for speci c indications as the procedure is considered as novel and the Table 6. The relationship between post-swim-up sperm morphology and fertilization rates and pregnancy rates Post-swim-up normal sperm morphology >25% 25% No. of IVF cycles Eggs inseminated Eggs fertilized Fertilization rate 78.6 a 55.0 a No. of ET No. of pregnancy 35 2 Pregnancy rate/et (%) a p < safety aspect is to a great extent unexplored. ICSI should not be used in a rst cycle even if only few oocytes were obtained with an apparently normal sperm sample. When there are some reasons to suspect poor fertilization, it is acceptable that ICSI can be used in combination with conventional IVF in a split cycle. On the contrary, it was reported that the majority of IVF cycles with poor fertlization was because of transient factors, possibly temporary poor sperm samples or non-optimal stimulation protocols (Hamberger et al., 1998). In this study, sperm morphology was investigated before and after the sperm separation to predict unexpectedly poor fertilization outcome in patients without severe male infertility, such as `poor prognosis' as determined by the strict criteria, to avoid cancellation of ET by shifting from IVF to ICSI. As the evaluation of sperm morphology by strict criteria is a simple, rapid and cost-effective method, it is becoming a popular method for evaluating sperm function, at least in assisted reproduction (Kruger et al., 1988). In a recent review by Coetzee et al. (1998), the majority of the articles (36/ 43 ˆ 81.4%) concluded that normal sperm morphology, including acrosomal morphology, had a role to play in the diagnosis of male fertility potential. In the present study, the strict criteria were used to establish a relative indication for ICSI in patients without severe male infertility. When native spermatozoa were evaluated, sperm morphology demonstrated a signi cant correlation with the fertilization rate, which 4 supported the previous studies (Kruger & Coetzee, 1999). The fertilization rate (80.5%) in 110 `normal' samples was signi cantly higher than that (55.4%) in 27 `good prognosis' samples. No embryo was available in 3.6% of 110 `normal' cycles and in 11.1% of 27 `good prognosis' cycles. However, there was no difference between the two groups. These results indicate that the Kruger's strict criteria for the native spermatozoa provide a reliable estimation of the fertilizing ability of human spermatozoa. Some reports have demonstrated sperm functional problems associated with teratozoospermia. Sperm-zona pellucida binding was impaired when spermatozoa from men with teratozoospermia were used in the hemizona assay (Franken et al., 1990). The acrosome reaction is an exocytotic process involving fusion of the sperm plasma membrane and the outer acrosomal membrane. A low incidence of spontaneous Category n Sperm morphology Before swim-up (%) Post swim-up (%) Fertilization rate (%) (eggs fertilized/ inseminated) Pregnancy rate (%) (pregnancy/et) Table 7. Comparison of the fertilization and pregnancy rates between patients in Categories A and B A a (50/106) 11.1 (1/9) B > a (74/123) 35.7 (5/14) a p < 0.05.

6 Sperm morphology and fertilization 107 acrosome reaction and a diminished progesterone-stimulated acrosome reaction were observed in infertile patients with a high incidence of abnormal sperm forms according to the strict criteria (Oehninger et al., 1994). Sperm nuclear normality assessed using Aniline Blue stain has been reported to 5 be correlated with normal morphology (Liu & Baker, 1992). Sperm nuclear maturity diagnosed using Acridine Orange stain was also found to be moderately correlated with normal sperm morphology (Claassens et al., 1992). These problems associated with teratozoospermia could be overcome when ICSI was clinically introduced. Nagy et al. (1995) reported that no differences were noted between the three morphology groups (0, 1±3 and 14%) either in the fertilization rate, the embryo development rate, the transfer rate or the pregnancy rate when results of ICSI were analysed according to the morphological characteristics of the spermatozoa. A number of studies have shown that sperm separation techniques, including swim-up and Percoll gradient centrifugation, select spermatozoa with higher motility and better morphology (Lopata et al., 1976; Gorus & Pipeleers, 61981). Prakash et al. (1998) recently reported that the differential gradient sperm separation method using Percoll was superior to the swim-up method for selecting spermatozoa with normal morphology as assessed by the strict criteria. On the contrary, Smith et al. (1995) suggested that no single separation technique was always superior, and sperm specimens should be evaluated individually to determine the best separation method before actual ART cycles. There were two investigations that evaluated the use of sperm morphology assessment by the strict criteria on the post-swim-up (Enginsu et al., 1993) or post-percoll (Yue et al., 1995) separated spermatozoa used for oocyte insemination in IVF programmes. The former study concluded that the percentages of spermatozoa with normal morphology using strict criteria, both in native and in post-swim-up samples, were the best predictors of IVF outcome. They divided patients into three categories according to the normal morphology before and after swim-up separation and found that patients with one of the categories (native <5% and post-swim-up <8% normal forms) showed a 7% fertlization rate and a 0% pregnancy rate. The latter concluded that <4% normal spermatozoa in the Percolltreated sample predicted a lower cleavage rate and a signi cant but marginally reduced number of ET. In these studies, it is clear that treatment with ICSI should absolutely be performed according to their cut-off points. However, they did not direct their attention to the patients without severe male factor infertility and having unexpectedly poor fertilization outcome. In the present study, sperm were separated using the swim-up method for oocyte insemination. Even after separation, sperm morphology was signi cantly correlated with the fertilization rate in IVF±ET, while other sperm parameters were not. When the cut-off level for the post swim-up sperm morphology was set at 25%, there was a signi cant difference in the fertilization rates between patients (78.6%) with post-swim-up >25% and those (55.0%) with post swim-up <25%. Taken together, a relative indication for ICSI using the sperm morphology before and after swim-up was established and termed Categories A and B. There was a signi cant difference of the fertlization rates between patients in these two categories. Moreover, the pregnancy rate of 11.1% in Category A patients was very low when compared with that of 35.7% in Category B patients. These ndings suggest that poor fertilization and pregnancy outcome in patients in Category A can be predicted using the strict criteria for sperm morphology before and after sperm separation in IVF±ET. Thus, ICSI is required in Category A patients to achieve a higher fertilization and pregnancy rate. In conclusion, the strict criteria provide a reliable estimation of the fertilizing ability of human spermatozoa. The information obtained using sperm morphology assessments before and after sperm separation might be useful in planning strategies for the treatment of infertile couples. Acknowledgements We are grateful to Ms Chieko Yamaguchi (IVF Laboratory, Jichi Medical School) for her technical assistance. Alper, M. M., Lee, G. S., Seibel, M. M., Smith, D., Oskowitz, S. P., Ransil, B. J. & Taymar, M. L. (1985) The relationship of semen parameters to fertilization in patients participating in a program of in vitro fertilization. Journal of In Vitro Fertilization and Embryo Transfer 2, 217±223. Bonduelle, M., Legein, J., Buysse, A., Van Assche, E., Wisanto, A., Devroey, P., Van Steirteghem, A. C. & Liebaers, J. (1996) Prospective study of 877 children born after intracytoplasmic sperm injection. Human Reproduction 11, 1558±1564. Burkman, L. J., Coddington, C. C., Franken, D. R., Kruger, T. F., Rosenwaks, Z. & Hodgen, G. D. (1988) The hemizona assay References (HZA): development of a diagnostic test for the binding of human spermatozoa to human hemizona pellucida to predict fertilization potential. Fertility and Sterility 49, 688±697. Chan, S. Y., Wang, C., Chan, S. T., Ho, P. C., So, W. W., Chan, Y. F. & Ma, H. K. (1989) Predictive value of sperm morphology and movement characteristics in the outcome of in vitro fertilization of human oocytes. Journal of In Vitro Fertilization and Embryo Transfer 6, 142±148. Claassens, O. E., Menkveld, R., Franken, D. R., Pretorius, E., Swart, Y., Lombard, C. J. & Kruger, T. F. (1992) The Acridine Orange

7 108 H. Obara et al. test: determining the relationship between sperm morphology and fertilization in vitro. Human Reproduction 7, 242±247. Coetzee, K., Kruger, T. F. & Lombard, C. J. (1998) Predictive value of normal sperm morphology: a structured literature review. Human Reproduction Update 4, 73±82. Enginsu, M. E., Dumoulin, J. C. M., Pieters, M. H. E. C., Evers, J. L. H. & Geraedts, P. M. (1993) Predictive value of morphologically normal sperm concentration in the medium for in-vitro fertilization. International Journal of Andrology 16, 113±120. Franken, D. R., Kruger, T. F., Menkveld, R., Oehninger, S., Coddington, C. C. & Hodgen, G. D. (1990) Hemizona assay and teratozoospermia: increasing sperm insemination concentrations to enhance zona pellucida binding. Fertility and Sterility 54, 497±503. Gorus, F. K. & Pipeleers, D. G. (1981) A rapid method for the fractionation of human spermatozoa according to their progressive motility. Fertility and Sterility 35, 662±665. Grow, D. R., Oehninger, S., Seltman, H. J., Toner, J. P., Swanson, R. J., Kruger, T. F. & Muasher, S. J. (1994) Sperm morphology as diagnosed by strict criteria: probing the impact of teratozoospermia on fertilization rate and pregnancy outcome in a large in vitro fertilization population. Fertility and Sterility 62, 559±567. Hamberger, L., Lundin, K., Sjogren, A. & Soderlund, B. (1998) Indications for Intracytoplasmic sperm injection. Human Reproduction 13 (Suppl. 1), 128±133. Hirsch, I., Gibbons, W. E., Lipshultz, L. I., Rossarik, K. K., Young, R. L., Poindexter, A. N., Dudson, M. G. & Findley, W. E. (1986) In vitro fertilization in couples with male factor infertility. Fertility and Sterility 45, 659±664. Jeulin, C., Feneux, D., Serres, C., Jouannet, P., Guillet-Rasso, F., Belaisch-Allart, J., Frydman, R. & Testart, J. (1986) Sperm factors related to failure of human in vitro fertilization. Journal of Reproduction and Fertility 76, 735±744. Kruger, T. F., Menkveld, R., Stander, F. S., Lombard, C. J., Van der Merwe, J. P., Van Zyl, J. A. & Smith, K. (1986) Sperm morphologic features as a prognostic factor in in vitro fertilization. Fertility and Sterility 46, 1118±1123. Kruger, T. F., Acosta, A. A., Simmons, K. F., Swanson, R. J., Matta, J. F. & Oehninger, S. (1988) Predictive value of abnormal sperm morphology in in vitro fertilization. Fertility and Sterility 49, 112±117. Kruger, T. F. & Coetzee, K. (1999) The role of sperm morphology in assisted reproduction. Human Reproduction Update 5, 172±178. Liu, D. Y., Lopata, A., Johnston, W. I. H. & Baker, H. W. G. (1989) Human sperm-zona Pellucida binding, sperm characteristics and in vitro fertilization. Human Reproduction 4, 696±701. Liu, D. Y. & Baker, H. W. G. (1992) Sperm nuclear chromatin normality: relationship with sperm morphology, sperm-zona pellucida binding, and fertilization rates in vitro. Fertility and Sterility 58, 1178±1184. Lopata, A., Patullo, M. J., Chang, A. & James, B. (1976) A method for collecting motile spermatozoa from human semen. Fertility and Sterility 27, 677±684. Lucena, E., Lucena, C., Gomez, M., Ortiz, J. A., Ruiz, J., Arango, A., Diaz, C. & Benerman, C. (1989) Recovery of motile sperm using the migration-sedimentation technique in an in-vitro fertilization-embryo transfer programme. Human Reproduction 4, 163±165. Mahadevan, M. M. & Trounson, A. O. (1984) The in uence of seminal characteristics on the success rate of human in vitro 7 fertilization. Fertility and Sterility 42, 400±405. Nagy, Z. P., Liu, J., Joris, H., Verhexen, G., Tournaye, H., Camus, M., Derde, M. C., Devroey, P. & Van Steirtegem, A. C. (1995) The result of intracytoplasmic sperm injection is not related to any of the three basic sperm parameters. Human Reproduction 10, 1123±1129. Oehninger, S., Blackmore, P., Morshedi, M., Sueldo, C., Acosta, A. A. & Alexander, N. J. (1994) Defective calcium in ux and acrosome reaction (spontaneous and progesterone-induced) in spermatozoa of infertile men with severe teratozoospermia. Fertility and Sterility 61, 349±354. Prakash, P., Leykin, L., Chen, Z., Toth, T., Sayegh, R., Schiff, I. & Isaacson, K. (1998) Preparation by differential gradient centrifugation is better than swim-up in selecting sperm with normal morphology (strict criteria). Fertility and Sterility 69, 722±726. Shibahara, H., Hamada, Y., Hasegawa, A., Toji, H., Shigeta, M., Yoshimoto, T., Shima, H. & Koyama, K. (1999) Correlation between the motility of frozen-thawed epididymal spermatozoa and the outcome of intracytoplasmic sperm injection. International Journal of Andrology 22, 324±328. Shibahara, H., Mitsuo, M., Inoue, M., Hasegawa, A., Shigeta, M. & Koyama, K. (1998) Relationship between human in-vitro fertilization and intracytoplasmic sperm injection and the zonafree hamster egg penetration test. Human Reproduction 13, 1928± Shibahara, H., Naito, S., Hasegawa, A., Mitsuo, M., Shigeta, M. & Koyama, K. (1997) Evaluation of sperm fertilizing ability using the Sperm Quality Analyzer. International Journal of Andrology 20, 112±117. Smith, S., Hosid, S. & Scott, L. (1995) Use of postseparation sperm parameters to determine the method of choice for sperm preparation for assisted reproductive technology. Fertility and Sterility 63, 591±597. Van Steirteghem, A., Liu, J., Joris, H., Nagy, Z., Janssenswillen, C., Tournaye, H., Derde, M. P., Van Assche, E. & Devroey, P. (1993) Higher success rate by intracytoplasmic sperm injection than by subzonal insemination. Report of a second series of 300 consecutive treatment cycles. Human Reproduction 8, 1055±1060. Yanagimachi, R., Yanagimachi, H. & Rogers, B. J. (1976) The use of zona-free animal ova as a test system for the assessment of fertilization capacity of human spermatozoa. Biology of Reproduction 15, 471±476. Yovich, J. L. & Stanger, J. D. (1984) The limitations of in vitro fertilization from males with severe oligospermia and abnormal sperm morphology. Journal of In Vitro Fertilization and Embryo Transfer 1, 172±179. Yue, Z., Meng, F. J., Jorgensen, N., Ziebe, S. & Andersen, A. N. (1995) Sperm morphology using strict criteria after Percoll density separation; in uence on cleavage and pregnancy rates after in-vitro fertilization. Human Reproduction 7, 1781± Received 6 October 2000; accepted 20 December 2000