Successful pregnancy in a case of azoospermic infertility by using testicular sperm for intracytoplasmic injection into the oocyte

Transcription

1 Intracytoplasmic sperm injection Successful pregnancy in a case of azoospermic infertility by using testicular sperm for intracytoplasmic injection into the oocyte MKH Leong, KK Wong, CKM Leung, C Wong, W Tang, F Tang Non-obstructive azoospermia used to be considered an untreatable cause of infertility. By the microinjection technique, however, sperm that has been surgically extracted from the testis can be injected into the oocyte cytoplasm. The injected eggs can be transferred to the uterus or fallopian tubes to initiate a pregnancy. A healthy baby boy conceived by using this method was delivered in November This micromanipulation technique offers couples in which the man has non-obstructive azoospermia the chance of having their own offspring. The methodology used and a brief discussion of its merits are presented. HKMJ 1999;5:69-71 Key words: Fertilization in vitro; Microinjections; Oligospermia; Reproduction techniques Introduction In 1978, the world s first baby conceived by the in vitro fertilisation (IVF) technique, Louise Brown, was born. Since then, advancements have been made in the IVF technique and its allied technologies. Of all advancements, the most significant are the micromanipulation techniques, which, when used singly or in combination, are instrumental in circumventing male fertility problems that had been considered insurmountable. Intracytoplasmic sperm injection (ICSI) was first described in and has revolutionised the management of severe male infertility; severe asthenospermia, severe oligospermia, failed vasovasostomy, and azoospermia have all been successfully treated by using the ICSI technique. 2,3 This report is on a case of non-obstructive azoospermia that was successfully treated by a combination of IVF technologies in a private IVF programme in Hong Kong. Case report A 32-year-old man and 30-year-old woman who had no significant family or medical problems presented IVF Centre, Hong Kong Sanatorium and Hospital, 6th Floor, 2-4 Village Road, Happy Valley, Hong Kong MKH Leong, FRCOG, FHKAM (Obstetrics and Gynaecology) KK Wong, FRACS, FHKAM (Surgery) CKM Leung, FRCOG, FHKAM (Obstetrics and Gynaecology) C Wong, BSc W Tang, BA F Tang, BSc Correspondence to: Dr MKH Leong to the IVF Centre at the Hong Kong Sanatorium and Hospital in May The man was found to be azoospermic from two different semen analyses, and physical examination showed normal secondary sexual characteristics; the testes, epididymides, and vasa were normal. The level of follicle-stimulating hormone was 26.3 IU/L (normal range, IU/L). A testicular biopsy showed mostly Sertoli s cells and in the seminiferous tubules, only approximately 10% showed evidence of spermatogenesis. The couple was counselled about and subsequently consented to the ICSI technique and the need to recover sperm by performing a testicular biopsy. The first pregnancy In September 1996, controlled ovarian hyperstimulation was achieved by giving the woman intramuscular urofollitrophin (Metrodin; Serono Laboratory, Aubonne, Switzerland) 150 U twice daily, starting on day 3 of the menstrual cycle, for 6 days. Follicular growth was monitored by determining the plasma oestradiol level and by intravaginal ultrasonography. On day 10 of the menstrual cycle, human chorionic gonadotrophin U was given by intramuscular injection and oocyte collection was scheduled for 36 hours later (day 12). On the morning of day 12, an open testicular exploration/biopsy under general anaesthesia was performed on the man by a urologist. The biopsy tissue was transferred to a petri dish containing N-[2-hydroxyethyl]piperazine-N-[2-ethanesulphonic HKMJ Vol 5 No 1 March

2 Leong et al acid] (HEPES)-buffered Earle s medium (Sigma, St Louis, US) and shredded into small pieces using two 1-mL tuberculin syringes. The suspension was then examined under the microscope and when sperm were identified, the biopsy tissue was removed and the suspension was centrifuged at 300 g for 5 minutes. After the sperm pellet was resuspended, approximately 100 live, normal-looking mature sperm were identified. Oocyte collection was performed by a gynaecological surgeon while the woman was sedated. By using an ultrasound-guided transvaginal needle aspiration technique, 10 oocytes were obtained. They were examined and were all found to be in the metaphase II stage of meiosis and suitable for ICSI. During the ICSI procedure, the woman was given anaesthesia and laparoscopy was commenced. When the ICSI was completed, six injected oocytes three to each ampulla were transferred to the fallopian tubes under video guidance. Pregnancy was confirmed 16 days after the procedure. The pregnancy progressed but spontaneous abortion occurred on the 45th day of gestation. The second pregnancy In March 1997, a second round of ovarian hyperstimulation was done. A second open testicular biopsy yielded only six live, mature, normal-looking spermatozoa. Transvaginal aspiration obtained six oocytes, five of which were in the metaphase II stage of meiosis. Following ICSI, two and three sperm-injected oocytes were transferred into the left and right fallopian tube, respectively. Pregnancy was confirmed 16 days after the procedure and progressed normally. At 17 weeks of gestation, an amniocentesis was conducted and showed the baby to be a normal male. Pregnancy progressed without complication, and a healthy baby boy, in footling breech presentation, was delivered by caesarean section in November Discussion Azoospermia used to be regarded as absolute sterility, and the available options were to adopt children or to artificially inseminate oocytes with donor sperm (donor insemination). The liberalisation of abortion laws, however, has made adoption more difficult. In addition, donor insemination has associated problems of a low availability of donor sperm, poor semen quality, and a possibility of transferring human 70 HKMJ Vol 5 No 1 March 1999 immunodeficiency virus. In Hong Kong, semen donors are scarce, and the long waiting lists and age limit imposed on recipients in the local donor insemination programme effectively rule out many couples from the programme. Experience around the world has shown that in most men who present with non-obstructive azoospermia for testicular sperm extraction, the testes have islets of normal spermatogenesis in the seminiferous tubules and thus contain normal sperm. 2-4 Unfortunately, the number of normal sperm is usually insufficient for IVF, which requires motile sperm. Allied procedures are thus needed. Intracytoplasmic sperm injection has brought hope to a situation where the wife is producing more eggs than the husband s spermatozoa situations never even contemplated in our wildest dreams 2 to 3 years ago. 5 Such situations occurred on this couple s second attempt at pregnancy, when only six usable sperm were identified. Micromanipulation followed by immediate fallopian tube transfer (MIFT) was first reported in 1994 by McLachlan et al, 6 who replaced oocytes after performing subzonal sperm injection. We believe that ICSI followed by immediate fallopian tube transfer is a better procedure, because there is a higher fertilisation rate. Treating infertility by using assisted reproduction techniques such as gamete (sperm and egg) intrafallopian transfer (GIFT) may be better than using IVF, because of the more physiological tubal environment for fertilisation and embryo development. 6 For the same reason, MIFT has a higher pregnancy rate than does ICSI-IVF. 6,7 Because of the lower fertilisation rate we have achieved with ICSI (IVF fertilisation rate: 75% of oocytes; ICSI fertilisation rate: 55% of oocytes; unpublished data), we compensate by placing slightly more oocytes than is usual in cases of unexplained infertility. Sperm retrieval can be achieved by open testicular biopsy or by epididymal aspiration. 2,4 A technique that immediately cryopreserves sperm-bearing testicular biopsies has recently been reported 8 and consequently, ICSI-IVF or MIFT can be carried out at a later, more convenient date. This method also obviates the logistic difficulty of conducting concurrent operative procedures on the couple. Conclusion We report on two pregnancies in a couple where the husband has non-obstructive azoospermia. In both attempts, using testicular sperm extraction and a direct

3 Intracytoplasmic sperm injection ICSI technique, coupled with immediate fallopian tube transfer, the woman became pregnant. The first pregnancy resulted in spontaneous abortion, but the second resulted in the delivery of a healthy, normal baby boy in November Testicular sperm retrieval and assisted reproduction techniques may now enable such couples to conceive. References 1. Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 1992;340: Silber SJ, Nagy Z, Lin J, et al. The use of epididymal and testicular spermatozoa for intracytoplasmic sperm injection: the genetic implications for male infertility. Hum Reprod 1995;10: Devroey P, Liu J, Nagy Z, et al. Pregnancies after testicular sperm extraction and intracytoplasmic sperm injection in nonobstructive azoospermia. Hum Reprod 1995;10: Craft I, Tsirigotis M, Bennett V, et al. Percutaneous epididymal sperm aspiration and intracytoplasmic injection in the management of infertility due to obstructive azoospermia. Fertil Steril 1995;63: Edwards RC. Human conception in vitro 1995, a summing up. In: Edwards RC, Beard HK, Howles CM, editors. Human Conception in vitro Oxford: Oxford University Press; 1996: McLachlan RI, Fuscaldo G, Calderon I, et al. Microinjection: choice of embryo transfer technique. Reprod Fertil Dev 1994; 6: Meirow D, Schenker JG. Appraisal of gamete intrafallopian transfer. Eur J Obstet Gynecol Reprod Biol 1995;58: Salzbrunn A, Benson DM, Holstein AF, Schulze W. A new concept for the extraction of testicular spermatozoa as a tool for assisted fertilization (ICSI). Hum Reprod 1996;11: Editorial comment Multiple pregnancy is an unwanted outcome of assisted reproductive technology and is associated with a relatively high perinatal morbidity and mortality. Measures are now being taken in some countries to reduce the numbers of multiple pregnancies resulting from these procedures. In the United Kingdom, the maximum number of embryos that can be transferred to patients, by law, is three. The American Society for Reproductive Medicine suggests that no more than three good embryos be transferred to women aged under 35 years, no more than four good embryos to women aged 35 to 40 years, and no more than five good embryos to women older than 40 years or who have had multiple failed attempts. 1 There is currently no legislation in Hong Kong regarding the number of embryos that can be transferred. Reference 1. American Society for Reproductive Medicine. Guidelines on number of embryos transferred. A Practice Committee Report, Website embryos.htm HKMJ Vol 5 No 1 March

4 Human Reproduction vol.14 no.12 pp , 1999 Testicular sperm extraction: impact of testicular histology on outcome, number of biopsies to be performed and optimal time for repetition Medhat Amer 1,2,3, Shawky El Haggar 1, Taymour Moustafa 1, Taha Abd El-Naser 1,2 and Wael Zohdy 1 1 Department of Andrology, Cairo University Hospitals and 2 Adam International Clinic, 20 Aden Street, Mohandessin, Giza, Egypt 3 To whom correspondence should be addressed active spermatogenesis from which a tiny number of spermatozoa can be extracted (Silber et al., 1995). That is why most investigators prefer open biopsies (Friedler et al., 1997; Ezeh et al., 1998; Rosenlund et al., 1998) or biopty gun needle biopsy (Rajfer and Binder, 1989; Tuuri et al., 1998; Hovatta et al., 1999) rather than needle aspiration (Craft et al., 1997) to maximize chances of finding these rare foci of active spermatogenesis. As there is inconsistency in the literature concerning the pattern of testicular histology in non-obstructive azoospermic men (Hauser et al., 1998), attitudes vary con- cerning the number of testicular samples that should be taken for sperm retrieval in men with non-obstructive azoospermia. Some perform a single testicular biopsy (Silber et al., 1995; Verheyen et al., 1995) based on the assumption that multi- focal distribution of spermatogenesis throughout the entire testis is present in non-obstructive azoospermia (Silber et al., 1997). Others perform and strongly recommend multiple biopsies, since this may enhance the diagnostic accuracy of absolute testicular failure and increase the number of sperm cells retrieved (Tournaye et al., 1996, 1997; Hauser et al., 1998; Ostad et al., 1998). If the multiple surgical approach is preferred, there is an increased risk of interruption of the blood supply (Schlegel and Su, 1997), post-sampling fibrosis or autoimmune response (Tournaye et al., 1997). As the detri- mental effect of TESE on spermatogenesis continues for several months, it was advised (Schlegel and Su, 1997) not to repeat the procedure on the same testis within 6 months. Testicular sampling evokes two important questions: the optimal number of biopsies that should be performed and the optimum time for repetition. The first aim of this study was to answer these two important questions by determining the percentile incidence of successful TESE in patients with functional azoospermia in relation to various histopathological patterns and to the number of performed biopsies. This is important, as it helps to decide in which histopathological pattern the performance of multiple, rather than a single, testicular sample can be more beneficial and to avoid unnecessary multiple sampling. Our second aim was to evaluate the likelihood of success or failure of second TESE in relation to testicular function and time passed after the first procedure, thus determining the optimal time to repeat TESE procedure if needed. Testicular sperm extraction (TESE) may not always be successful in patients with non-obstructive azoospermia, as they only have minute foci of active spermatogenesis from which a tiny number of spermatozoa can be extracted. The aim of this study was to find the percentile incidence of successful TESE in non-obstructive azoospermia patients in relation to various histopathological patterns and the number of performed biopsies, and to determine the optimal time needed for repetition. A total of 216 patients underwent bilateral testicular biopsy taking a single piece from each testis for sperm retrieval and pathological evaluation. In another 100 patients, the same procedure was done but taking multiple samples (maximum four samples/testis). Spermatozoa were successfully retrieved from 37.5 and 49% of patients who supplied single and multiple samples respectively. TESE was significantly higher when multiple samples were taken in all histopathological groups except for Sertoli cell-only syndrome, tubular sclerosis and Klinefelter s pattern. Twenty-seven patients underwent repeated TESE for ICSI between 1 and 24 months from the first procedure; all of them had easy sperm retrieval during the first procedure. Although sperm retrieval was successful in 75 and 94.7% of patients who underwent the second attempt, before and after 3 months respectively, a second TESE was usually more difficult and necessitated multiple sampling. Key words: azoospermia/sperm retrieval/testicular biopsy/ testicular sperm extraction Introduction It is well established that mature testicular spermatozoa could be found in men with non-obstructive azoospermia (Hauser et al., 1994; Tournaye et al., 1995). Thus, testicular sperm extraction (TESE) combined with intracytoplasmic sperm injection (ICSI) offer azoospermic patients the possibility of fathering their own genetic children even if they do not have normal spermatogenesis. TESE and ICSI have become the standard protocols for treatment of patients with functional azoospermia who have no source of spermatozoa except the testis (Devroey et al., 1995). However, TESE may not always be successful in those patients who have only minute foci of Materials and methods Two groups (A and B) of patients with azoospermia were recruited from the Adam International Clinic and the Cairo University Andrology Outpatient Clinic. Group A included 316 patients with functional azoospermia: it 3030 European Society of Human Reproduction and Embryology

5 Human Reproduction vol.13 no.1 pp , 1998 Multiple pregnancies obtained by testicular spermatid injection in combination with intracytoplasmic sperm injection S.Kahraman 1,3, G.Polat 1, M.Samli 2, E.Sözen 1, O.D.Özgün 1, K.Dirican 1 and T.Özbiçer 1 1 Assisted Reproductive Techniques and Reproductive Endocrinology Unit and 2 Urology Department, Sevgi Hospital, Ankara, Turkey 3 To whom correspondence should be addressed at: Department of Reproductive Endocrinology and Infertility, Tunus Caddesi, No: 28, Kavaklidere, Ankara, 06680, Turkey Recent studies have shown that the injection of spermatid cells into the human oocyte can result in normal fertilization, embryo development and even delivery of live, healthy offspring. In our study, 23 azoospermic cases with severe spermatogenetic defects in their testicular biopsy are presented. The serum follicle stimulating hormone (FSH) concentrations and histopathological results of these males have been documented and compared in terms of fertilization and embryo development. The mean FSH value of the azoospermic males was miu/l, ranging from 1.6 to 39 miu/l. Elongated spermatids were used in three cases only, as these more mature forms were mostly present in the testicular sample. In the remaining 20 cases, only round spermatids were found for use in intracytoplasmic sperm injection (ICSI). The fertilization rate with two pronuclei was 31.3%. The fertilization rate was found to be as high as 71% in three patients in the elongating and elongated spermatids group and as low as 25.6% in the round spermatid group. A few immature, non-motile spermatozoa were seen in only two cases from the elongated spermatid group. However, in the remaining cases, no spermatozoa were observed. The number of pronuclear (PN) arrest was quite high when only round spermatids were used (36.1%). Total fertilization failure was observed in two cases from the round spermatid group with Sertoli cell only and germ cell aplasia. A total of three pregnancies was achieved in 23 cases (13.0%), two from the elongated spermatid group and one from the round spermatid group. One biochemical pregnancy with a round spermatid resulted in an early spontaneous abortion and surprisingly, the remaining pregnancies were achieved with elongated spermatids resulting in multiple pregnancies. One twin and one triplet pregnancy were established following four embryo transfers in each patient. The twin pregnancy resulted in a live birth with two healthy babies; unfortunately, the triplet pregnancy ended in an abortion at 11 weeks. The use of testicular spermatids in the treatment of non-obstructive azoospermia may give hope by offering a novel treatment model. In cases with very severe spermatogenetic defect, even multiple pregnancies can be achieved with elongated spermatid cells by yielding a high implantation rate. However, the efficiency of round spermatids in achieving fertilization and pregnancy was disappointing. Key words: intracytoplasmic sperm injection/multiple pregnancy/testicular spermatid injection Introduction Although it is understood in animal studies that fertilization and pregnancy can be achieved using round spermatids and even with secondary spermatocytes, implementation of this technique in human beings is a very novel concept (Kimura et al., 1995a,b). The ability of spermatid cells to fertilize a mature oocyte with elongated or round spermatid injection (ELSI, ROSI) procedures and delivery of live healthy offspring have been shown in several animal studies in both mouse and rabbit models (Ogura, 1993, 1994; Sofikitis, 1994a,b; Ogura and Yanagimachi, 1995). Successful fertilization of a human oocyte by a late stage spermatid with intracytoplasmic sperm injection (ICSI) was first reported by Vanderzwalmen (1995). The first two live births after intrauterine transfer of embryos obtained by injection of spermatids from semen were reported by Tesarik et al. (1996). Fishel et al. (1996) reported the first live birth after intrauterine transfer of embryos accomplished by utilization of testicular elongated spermatids. These pregnancies introduced the concept of using spermatids to alleviate severe cases of male infertility. Sofikitis (1994b) was the first to report a pregnancy achieved with an injection of round spermatids into rabbit oocytes resulting in a complete gestation. Spermatogenetic arrest is an interruption of the complex process of germ cell differentiation leading to the formation of spermatozoa, and it may result in either oligospermia (partial arrest) or azoospermia (total arrest). Various factors play a role in the aetiology of spermatogenetic arrest. Varicocele and bilateral cryptorchidism are the most common causes of nonobstructive azoospermia. Other factors such as infections and hormonal disturbances (gonadotrophin insufficiency or late onset adrenogenital syndrome, etc.) should be scrutinized in cases of incomplete arrest, as such conditions are treatable. Some pre-testicular causes may be due to hepatic and renal dysfunction, exposure to heat and toxic substances (chemotherapy, alcohol, some antibiotics) and nutritional disturbances (zinc, vitamin A, etc.). In cases of complete arrest, unless these factors are present, the arrest is usually at the primary 104 European Society for Human Reproduction and Embryology

6 S.Kahraman et al. Ovarian stimulation Gonadotrophin releasing hormone analogue (Suprefact ; Hoechst, Frankfurt, Germany) was given as a nasal spray, starting in the luteal phase, and was continued until the third day of menses, which was regarded as a sign of sufficient down-regulation. Follicular development was then stimulated with an injection of follicle stimulating hormone (FSH) (Metrodin ; Serono, Rome, Italy) and human menopausal gonadotrophin (HMG) (Humegon ; Organon, Oss, The Netherlands). Luteinization was induced by IU HCG (Pregnyl ; Organon, Istanbul, Turkey) injection and oocyte aspiration was performed 36 h later under guidance of vaginal ultrasound. Oocyte preparation After oocyte retrieval, the cumulus cells and the corona radiata were removed by a brief exposure to HEPES buffered Earle s medium containing 80 IU/ml hyaluronidase (type VIII, specific activity 320 IU/mg, H 3757 ; Sigma Chemical Co.). A cumulus mass was removed completely by aspiration of the cell complex in and out of a handdrawn glass Pasteur pipette. The oocytes were rinsed several times in IVF 50 and were assessed at 200 magnification under an inverted microscope to determine their stage of nuclear maturity and polar body existence. Oocytes were incubated in IVF 50 at 37 C in an atmosphere of 5% CO 2 in air covered by paraffin oil (Ovoil 150 ; Scandinavian IVF Science). ICSI (ELSI, ROSI) procedure The spermatids were aspirated from the pellet or Percoll suspension with a 10 µm pipette and transferred into the injection dish (Falcon 1006 ) containing a droplet of 5 µl Gamete 100 medium and kept in the incubator until the ICSI procedure. They were then aspirated one by one from the droplet and transferred into the polyvinylpyrrolidone droplet (ICSI 100 ; Scandinavian IVF Science). A 7 µm micropipette was used for ICSI. In the case of late spermatids, no mid-piece could be observed and the beginning of the tail was very rigid when touched. The head of the round spermatid is compressed and takes on an oval shape when entering the pipette during the ICSI procedure as previously described by Vanderzwalmen et al. (1995). The spermatid was injected into the oocyte after vigorous aspiration of ooplasm to facilitate activation. After intracytoplasmic injection, spermatids can be followed clearly within the ooplasm due to their large size. Injected oocytes were placed into individual droplets in a Petri dish. Further evaluation of injected oocytes and establishment of pregnancy Fertilization was checked at 16, 18 and 24 h after injection. Fertilization was assessed as normal when two clearly distinct pronuclei containing nuclei were present. The state of embryo cleavage and quality was assessed after a further 24 h of in-vitro culture. The embryos were evaluated according to the blastomere size equality and the relative proportion of anucleate fragments. Figure 3 represents middle and low quality embryos attained by spermatid injection. One pregnancy was achieved with these mediocre quality embryos with a round spermatid injection (ROSI); unfortunately, this pregnancy ended in a spontaneous abortion. A maximum of four embryos was transferred, and assisted hatching was applied to each embryo using acidified Tyrode s solution. All patients underwent an initial test for serum β-hcg on day 12 after embryo transfer. Clinical pregnancy was diagnosed by ultrasonography at 7 weeks of gestation. Statistical analysis Fisher s exact test was used to compare FSH concentrations with fertilization and pregnancy. No statistical evaluation of the variation 106 Figure 3. Low (right) and medium (left) quality embryos obtained with round spermatids demonstrating uneven blastomeres and fragmentation. Table I. Distribution of follicle stimulating hormone (FSH) concentrations, testicular biopsy results and infertility aetiology of azoospermic males Patient FSH (miu/ml) Results of testicular biopsy Aetiology MA PTF 2 39 MA, HS Mumps 3 7 MA, HS PTF 4 12 Complete MA PTF Complete MA PTF Complete MA PTF 7 11 GCA, MA PTF 8 28 Complete MA PTF 9 23 GCA Cryptorchidism GCA PTF Complete MA PTF MA, HS PTF MA, HS PTF MA, HS PTF MA, HS PTF SCO PTF SCO PTF Complete MA PTF HS PTF GCA PTF HS PTF Complete MA PTF 23 7 SCO PTF Mean Range MA maturation arrest, HS hypospermatogenesis, GCA germ cell aplasia, SCO Sertoli cell only syndrome, PTF primary testicular failure. in fertilization rates of elongated and round spermatids was possible because of the low numbers involved. Results The mean ages of females and azoospermic males were and years respectively. The mean duration of infertility was years. The mean testicular volume was found to be ml. Serum FSH levels, biopsy results and aetiology distribution of 23 azoospermic cases are presented in Table I. In most cases (n 21) primary testicular failure was diagnosed. Mumps orchiditis was diagnosed in one

7 The New England Journal of Medicine Brief Report BIRTHS AFTER INTRACYTOPLASMIC INJECTION OF SPERM OBTAINED BY TESTICULAR EXTRACTION FROM MEN WITH NONMOSAIC KLINEFELTER S SYNDROME GIANPIERO D. PALERMO, M.D., PETER N. SCHLEGEL, M.D., E. SCOTT SILLS, M.D., LUCINDA L. VEECK, M.L.T., NIKICA ZANINOVIC, M.SC., SILVIA MENENDEZ, M.SC., AND ZEV ROSENWAKS, M.D. KLINEFELTER S syndrome is a form of hypergonadotropic hypogonadism and infertility resulting from a supernumerary X chromosome (47,XXY), with an incidence of approximately 1 case in 500 phenotypic males. 1,2 Some men with Klinefelter s syndrome who have chromosomal mosaicism (46,XY/47,XXY) are fertile. Men with nonmosaic, or complete, Klinefelter s syndrome usually have azoospermia, and only a few have any spermatogenesis. 3,4 Intracytoplasmic sperm injection, in which a spermatozoon is injected into an ovum in vitro, is an effective treatment for male-factor infertility. However, the complete absence of spermatozoa presents a particular clinical challenge. Postorchitis atrophy and genetic anomalies are the main causes of nonobstructive azoospermia, which is characterized by germ-cell aplasia, maturation arrest, or hypospermatogenesis. In such cases, testicular extraction of sperm has proved useful in obtaining sufficient sperm for fertilization with intracytoplasmic sperm injection. 5,6 We report on two couples, in each of which the man had nonmosaic Klinefelter s syndrome, who underwent testicular sperm extraction and intracytoplasmic sperm injection, which resulted in the delivery of healthy infants. Couple 1 CASE REPORTS A healthy 32-year-old man and a healthy 32-year-old woman were evaluated after two years of primary infertility. A testicularbiopsy specimen obtained from the man one year earlier contained only Sertoli cells. He had a gynecoid habitus, scant facial hair, slight gynecomastia, bilaterally atrophic testes, and small bi- From the Center for Reproductive Medicine and Infertility, Department of Obstetrics and Gynecology (G.D.P., E.S.S., L.L.V., N.Z., S.M., Z.R.), and the James Buchanan Brady Foundation, Department of Urology (P.N.S.), New York Hospital Cornell Medical Center, New York. Address reprint requests to Dr. Palermo at HT-336, New York Hospital, 505 E. 70th St., New York, NY , Massachusetts Medical Society. lateral varicoceles. His serum gonadotropin concentrations were high, and his serum testosterone concentration was low. Analysis of three semen specimens showed normal volumes and fructose concentrations; a single, abnormal, nonmotile sperm was seen in one specimen. On the basis of an analysis of 50 peripheral-blood leukocytes, his karyotype was 47,XXY. The woman was normal. We performed a single in vitro fertilization cycle with intracytoplasmic injection of sperm obtained by testicular extraction. During this cycle, the woman was treated first with a gonadotropin-releasing hormone agonist (leuprolide), administered subcutaneously, to inhibit gonadotropin secretion and then with a combination of human menopausal gonadotropin and folliclestimulating hormone, administered intramuscularly, to stimulate the development of ovarian follicles. Fifteen oocytes were retrieved approximately 34 to 36 hours after chorionic gonadotropin had been given intramuscularly. Ten oocytes were fertilized, and three were transferred into the uterus, but no pregnancy resulted. No embryos were considered suitable for cryopreservation. For the second in vitro fertilization cycle, which was performed six months later, the woman received leuprolide and follicle-stimulating hormone. After chorionic gonadotropin had been administered, 13 oocytes were retrieved by ultrasonographically guided transvaginal needle aspiration; 12 were at the second stage of metaphase. Bilateral testicular biopsy to obtain sperm was performed at the same time. 6 Before undergoing the biopsy, the man had received testolactone (100 mg orally twice daily) for three months. The biopsy specimens yielded 600 mg of seminiferous tubules. On microscopical examination, the tubules contained mostly Sertoli cells and only a few spermatogenic elements; extensive Leydig-cell hyperplasia was also seen (Fig. 1). A search of fresh tissue under higher magnification identified approximately 10 sperm with twitching movements only. Enough sperm were obtained by centrifugation of the tissue to fertilize each of the 12 ova by injection of a spermatozoon immobilized by crimping the tail. 5 Eight oocytes were normally fertilized, as indicated by the presence of two pronuclei and two extruded polar bodies. The couple declined preimplantation genetic testing. Because a proportion of the blastomeres were fragmented ( 20 percent) in the three embryos selected for transfer, assisted hatching (creation of an artificial breech in the zona pellucida) was performed according to previously reported methods. 7 The embryos were transferred into the uterus three days after fertilization. No embryos were cryopreserved. The woman was given intramuscular progesterone in oil (50 mg daily) until fetal cardiac activity was confirmed by ultrasonography. 7 Serum concentrations of the beta subunit of chorionic gonadotropin, measured twice weekly, increased to 94 miu per milliliter on day 11 after the embryo transfer. Ultrasonography performed 32 days later revealed two asymmetric intrauterine sacs, of which one had a fetal heartbeat. Amniocentesis at 20 weeks gestation showed a fetal karyotype of 46,XY. The pregnancy was normal, and the woman delivered a healthy 2778-g boy at 38.5 weeks gestation. Couple 2 A healthy 34-year-old man and a healthy 33-year-old woman were evaluated after five years of primary infertility. The man had a gynecoid habitus, bilaterally atrophic testes, gynecomastia, and a moderate-size left varicocele. Analysis of three semen samples showed low volume (1.2 ml) and normal fructose concentrations but no sperm. His serum gonadotropin concentrations were high, and his serum testosterone concentration was low. Bilateral testicular-biopsy specimens obtained approximately one year earlier contained only Sertoli cells. On the basis of an analysis of 40 peripheral-blood leukocytes, his karyotype was 47,XXY. The woman was normal. The woman was given leuprolide, human menopausal gonadotropin, and follicle-stimulating hormone. Forty mature oocytes were retrieved by transvaginal ultrasonography after the administration of chorionic gonadotropin. Simultaneous testicular biopsy, 588 February 26, 1998 The New England Journal of Medicine Downloaded from nejm.org on September 18, For personal use only. No other uses without permission. Copyright 1998 Massachusetts Medical Society. All rights reserved.

8 The New England Journal of Medicine one living spermatozoon to achieve fertilization. In otherwise normal men with nonobstructive azoospermia, small numbers of spermatozoa can sometimes be recovered by testicular biopsy and extraction for subsequent intracytoplasmic injection. Previous application of these techniques in men with nonmosaic Klinefelter s syndrome has resulted in fertilization and pregnancy, but not in birth Our results in two men indicate that this approach can be successful in men with nonmosaic Klinefelter s syndrome. Among men with nonobstructive azoospermia, including those with Klinefelter s syndrome, who agree to undergo testicular sperm extraction, some will be found to have no sperm. At present, there is no way to predict which men will have no sperm. 6 Couples should be prepared for the possibility that no sperm will be recovered and consider the use of donated sperm, if necessary. Testolactone (an aromatase inhibitor) is routinely given to men with nonobstructive azoospermia, including those with Klinefelter s syndrome (as in this study), to optimize semen production. During a three-year period, we have attempted testicular sperm extraction in 70 men with nonobstructive azoospermia, 6 of whom had either mosaic or nonmosaic Klinefelter s syndrome. This report describes the first two couples in which the man had nonmosaic Klinefelter s syndrome. Three of the four men in the other couples also had nonmosaic Klinefelter s syndrome, and spermatozoa were successfully retrieved in one of the four. With intracytoplasmic sperm injection, this couple conceived and now has an ongoing pregnancy. Although Klinefelter s syndrome is a nonheritable genetic condition that almost always results in sterility, meiosis is possible. However, spermatozoa from men with mosaic Klinefelter s syndrome contain an extra sex chromosome more often than do spermatozoa from normal men. 15,16 Therefore, there is a possibility that chromosomal errors will be transmitted to the offspring of men with Klinefelter s syndrome. The normal karyotypes of the three infants described here support previously reported data on the safety of intracytoplasmic sperm injection in the treatment of oligospermia and azoospermia. 17 As in all cases of severe male-factor infertility requiring intracytoplasmic sperm injection, genetic screening and prenatal testing should be strongly recommended. We are indebted to the clinical and scientific staffs of the Center for Reproductive Medicine and Infertility, and to Dr. J. Michael Bedford for his review of the manuscript. REFERENCES 1. Klinefelter HF Jr, Reifenstein EC Jr, Albright F. Syndrome characterized by gynecomastia, aspermatogenesis without A-Leydigism, and increased excretion of follicle-stimulating hormone. J Clin Endocrinol 1942;2: Paulsen CA, Plymate SR. Klinefelter s syndrome. In: King RA, Rotter JI, Motulsky AG, eds. The genetic basis of common diseases. Oxford, England: Oxford University Press, 1992: Paulsen CA, Gordon DL, Carpenter RW, Gandy HM, Drucker WD. Klinefelter s syndrome and its variants: a hormonal and chromosomal study. Recent Prog Horm Res 1968;24: Luciani JM, Mattei A, Devictor-Vuillet M, Rubin P, Stahl A, Vague J. Étude des chromosomes meiotiques dans un cas de maladie de Klinefelter avec spermatogenèse et caryotype 46,XY/47,XXY. Ann Genet 1970;13: Palermo GD, Schlegel PN, Colombero LT, Zaninovic N, Moy F, Rosenwaks Z. Aggressive sperm immobilization prior to intracytoplasmic sperm injection with immature spermatozoa improves fertilization and pregnancy rates. Hum Reprod 1996;11: Schlegel PN, Palermo GD, Goldstein M, et al. Testicular sperm extraction with intracytoplasmic sperm injection for nonobstructive azoospermia. Urology 1997;49: Palermo GD, Cohen J, Alikani M, Adler A, Rosenwaks Z. Intracytoplasmic sperm injection: a novel treatment for all forms of male factor infertility. Fertil Steril 1995;63: Rothwell NV. Sex chromosome anomalies in humans. In: Rothwell NV, ed. Understanding genetics: a molecular approach. New York: Wiley-Liss, 1993: Hargreave TB, ed. Male infertility. 2nd ed. London: Springer-Verlag, 1994:116-7, Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies after intracytoplasmic injection of a single spermatozoon into an oocyte. Lancet 1992;340: Devroey P, Nagy P, Tournaye H, Liu J, Silber S, Van Steirteghem A. Outcome of intracytoplasmic sperm injection with testicular spermatozoa in obstructive and non-obstructive azoospermia. Hum Reprod 1996;11: Harari O, Bourne H, Baker G, Gronow M, Johnston I. High fertilization rate with intracytoplasmic sperm injection in mosaic Klinefelter s syndrome. Fertil Steril 1995;63: Tournaye H, Staessen C, Liebaers I, et al. Testicular sperm recovery in nine 47,XXY Klinefelter patients. Hum Reprod 1996;11: Staessen C, Coonen E, Van Assche E, et al. Preimplantation diagnosis for X and Y normality in embryos from three Klinefelter patients. Hum Reprod 1996;11: Cozzi J, Chevret E, Rousseaux S, et al. Achievement of meiosis in XXY germ cells: study of 543 sperm karyotypes from an XY/XXY mosaic patient. Hum Genet 1994;93: Chevret E, Monteil M, Cozzi J, Pelletier R, Sèle B. Excess of hyperhaploid 24,XY spermatozoa in Klinefelter s syndrome detected by a threecolour FISH procedure. Fertil Steril 1995;64:Suppl:S235. abstract. 17. Palermo GD, Colombero LT, Schattman GL, Davis OK, Rosenwaks Z. Evolution of pregnancies and initial follow-up of newborns delivered after intracytoplasmic sperm injection. JAMA 1996;276: February 26, 1998 The New England Journal of Medicine Downloaded from nejm.org on September 18, For personal use only. No other uses without permission. Copyright 1998 Massachusetts Medical Society. All rights reserved.

9 ELSEVIER TESTICULAR SPERM EXTRACTION WITH INTRACYTOPLASMIC SPERM INJECTION FOR NONOBSTRUCTIVE AZOOSPERMIA PETER N. SCHLEGEL, GIANPIERO D. PALERMO, MARC GOLDSTEIN, SILVIA MENENDEZ, NIKICA ZANINOVIC, LUCINDA L. VEECK, AND ZEV ROSENWAKS ABSTRACT Objectives. To provide fertility for men with nonobstructive azoospermia. Methods. A retrospective review of treatment results at a university infertility center was undertaken. Sixteen couples entered an attempted in vitro fertilization (IVF)-intracytoplasmic sperm injection (ICSI) cycle for treatment of nonobstructive azoospermia. Each man was azoospermic, and the male factor diagnosis of nonobstructive azoospermia was made on testis biopsy for 14 men and on clinical grounds for 2 men. Sperm were retrieved by testicular biopsy on the day of oocyte retrieval. Results of testicular examinations, serum folliclestimulating hormone levels, and testicular histology as well as evaluation of the success rates of sperm retrieval, fertilizations, and pregnancies were made. Results. Sperm were extracted from testis biopsies in 10 of 16 (62%) testicular sperm extraction (TESE) attempts. For cycles in which sperm were retrieved, normal fertilizations were achieved for 51 of 98 (52%) mature oocytes injected with testicular sperm in 10 couples. Biochemical pregnancies were achieved for 6 of 16 (38%) couples, with clinical pregnancies during 5 of 16 (3 1%) attempts at sperm retrieval, and ongoing pregnancy and subsequent live delivery for 4 of 16 (25%) attempts. Conclusions. Pretreatment clinical parameters are unable to predict which men with nonobstructive azoospermia will have spermatozoa retrieved by TESE. When sperm are found, clinical pregnancies can occur for half (5/l 0) of these couples using TESE with ICSI, with ongoing pregnancy and delivery for 4 of 10 (40%). Many men with nonobstructive azoospermia will have retrievable sperm with testis biopsy that are suitable for ICSI; however, 6 of 16 (38%) couples will not have sperm retrieved with TESE and may undergo an unnecessary IVF procedure , ElsevierScience Inc. All rights reserved. UROLOGY 49: , T esticular failure affects approximately 1% of the male population and 10% of men who seek fertility evaluati0n.l Azoospermic men with testicular failure have either Sertoli cell-only pattern, maturation arrest, or hypospermatogenesis on testis biopsy. Until recently, it was assumed that men with nonobstructive azoospermia were untreatable. The only options offered to these couples to have children were the use of donor spermatozoa From thejames Buchanan Brady Foundation, The Department of Urology; Centerfor Reproductive Medicine and Infertility, The New York Hospital-Cornell Medical Center; and The Population Council, Center for Biomedical Research, New York, New York Reprint requests: Peter N. Schlegel, M.D., Department of Uralogy, Room F-905A, The New York Hospital-Cornell University Medical Center, 52.5 East 68th Street, New York, NY Submitted: April 25, 1996, accepted (with revisions): December COPYRIGHT 1997 BY ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED or adoption. Several clinically relevant findings have changed our approach to this condition. First, we have observed that direct evaluation of testis biopsy specimens often demonstrates sperm in men with nonobstructive azoospermia, despite severe defects in spermatogenesis.2 In addition, it was previously thought that sperm must traverse the male reproductive tract before acquiring the ability to normally fertilize an oocyte. Our experience with men who have unreconstructable obstructive azoospermia, including congenital absence of the vas deferens, suggests that complete transit through the epididymis is not a prerequisite for fertilization.3 Retrieval of sperm from the testis or epididymis was associated with high pregnancy rates using in vitro fertilization. Our subsequent experience indicated that micromanipulation of gametes during assisted reproduction could improve these pregnancy rates /97/$17.00 PII s (97)