To ICSI or Not to ICSI

Transcription

1 92 To ICSI or Not to ICSI GianpieroD.Palermo,MD,PhD 1 Queenie V. Neri, MSc 1 Zev Rosenwaks, MD 1 1 Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell Medical College, New York, New York Semin Reprod Med 2015;33: Address for correspondence Gianpiero D. Palermo, MD, PhD, Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell Medical College, 1305 York Avenue, Suite 720, New York, NY ( gdpalerm@med.cornell.edu). Abstract Keywords intracytoplasmic sperm injection male infertility in vitro fertilization spermatozoa When I (G.P.) was asked to write an article delineating the indications for intracytoplasmic sperm injection (ICSI), I confess that I was a little perplexed and hesitant. Being instrumentally involved in the appearance and increasing prevalence of this ART, I felt somewhat uneasy speaking for or against a broader use of this procedure. I think that the most unbiased thing to do is to explain how the need for ICSI first became apparent. During the sabbatical in Brussels that was my introduction to the world of in vitro fertilization (IVF), I was made aware very early on that total fertilization failure was thought to be the most frustrating aspect of in vitro insemination. This unpredictable event is particularly devastating for both the couple and the team treating them because it represented a seemingly impassable obstacle preventing a successful reproductive treatment. Intracytoplasmic sperm injection (ICSI) is the most effective assisted reproductive procedure enabling fertilization in severe forms of male factor indications and male gamete dysfunction. Reliability of ICSI has allowed the expansion of its application to other forms of infertility rendering it the most popular assisted reproduction technology (ART) insemination method worldwide. The concern related to the invasiveness of ICSI together with the arbitrary selection of the inseminating spermatozoon has induced the execution of studies to compare the performance of ICSI in non male factor infertility with standard in vitro insemination approach. Not surprisingly, the outcome has evidenced that ICSI does not yield higher pregnancy rates than in vitro fertilization but functions invariably as a normalizer of fertilization mollifying the absent or low fertilization. The follow-up studies on ICSI children have evidenced that the procedure is safe and the slightly higher incidences of neonatal malformations or de novo gonosomal abnormalities are related to the genetics of the infertile couples. Furthermore, ICSI is accepted for some specific indications such as low number and poor morphology oocytes, thicker zona, excess polyspermia, PGD/PGS/PGT (preimplantation genetic diagnosis/preimplantation genetic screening/preimplantation genetic testing), discordant HCV/HIV (hepatitis C virus/human immunodeficiency virus) couples, in vitro maturation (IVM), and oocyte cryopreservation. Only the advent of new biomarkers in combination with routine semen analysis capable of identifying the fertilization competence of the spermatozoon can guide the reproductive physician toward the proper insemination method. IVF entails the rigorous control and manipulation of the reproductive cycle aimed at recruiting and selecting a larger number of follicles. The desired number of oocytes is estimated and their quality evaluated, the spermatozoa are selected, and the number and quality of embryos transferred are arbitrarily determined. The only thing left to chance was the insemination itself, where everything is left to the natural intermingling of the two gamete types. After the establishment of IVF, it soon became clear that as many as 40% of the inseminated in vitro cycles were affected by fertilization failure or at best by an extremely low fertilization. 1 As a young investigator I was asked to address this issue and I subsequently embraced the male infertility aspect of the field, although I was initially reluctant due to my training in obstetrics and gynecology. My early work in Issue Theme BestPracticesinInVitro Fertilization; Guest Editor, Bradley J. Van Voorhis, MD Copyright 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) DOI /s ISSN

2 Intracytoplasmic Sperm Injection Palermo et al. 93 assisted fertilization involved subzonal injection (SUZI) of spermatozoa under the zona pellucida but outside the cytoplasmic membrane of the mouse oocyte was unable to withstand cytoplasmic injection. 2 After a few attempts performing the procedure in hamsters and unfertilized human oocytes, I was finally allowed to perform SUZI using gametes of actual patients. The couples chosen for the study had failed fertilization but still had an adequate number of spermatozoa. After the initial attempts by SUZI that yielded fertilization rates as high as 30.9%, I was then asked to inseminate sibling oocytes by SUZI and standard IVF to prove the efficacy of assisted fertilization. 2,3 I strongly disputed the use of both inseminations on the same oocyte cohort and asked instead that any male factor couple who were suitable candidates for assisted fertilization to be inseminated for as many attempts as needed to prove the inefficacy of the standard procedure. Once all conventional options were exhausted, I thought it would then be proper to use assisted fertilization by subzonal insemination. My motivation for this course of action was the fear that if embryos would be obtained through both insemination approaches, they would inevitably be replaced simultaneously during the same treatment cycle, clouding the benefit of the newly used technique. Therefore, it was agreed that a couple had to experience at least two complete fertilization failures with standard IVF to be eligible for the assisted fertilization protocol. The success in achieving fertilization by injecting an average of three spermatozoa beneath the zona was also due to the concurrent induction of the acrosome reaction, but even with this expedient, the fertilization outcome began to progressively decline when men with poorer quality and/or dysmorphic sperm were treated. 4 It was while performing SUZI that a single spermatozoon accidentally penetrated into the oolemma and provided the hint that a direct sperm injection would be more efficient. ICSI s effectiveness strongly suggested that any additional treatment of the spermatozoon was expendable, as ICSI enables the fertilization of most mature surviving oocytes. 5 7 The early use of ICSI required some adjustments to identify the best method to pierce the membrane and to identify the best location within the ooplasm in which to release the spermatozoon. It soon became apparent that this unconventional method of insemination was capable of fertilizing nearly every mature egg injected. Moreover, the ability to pinpoint the different steps of pronuclei appearance and to monitor the observation of the first embryonic cleavage without the obstructive layer of cumulus cells would facilitate tracking these relevant steps involved in the generation of the conceptus. In this work, we detail the current indications for ICSI and the array of applications of this insemination method to better gauge when it should or should not be used. The implications related to the use of ICSI in the dayto-day reproductive practice are discussed. Upcoming tests to help in deciding the proper insemination method to be implemented for a specific infertile couple are described. Popularity of Intracytoplasmic Sperm Injection According to various surveys, it appears that male factor infertility is commonly defined using the conventional semen profile, which provides descriptive information on the number of spermatozoa present in the ejaculate, the proportion that are motile or progressively motile, and the percentage of morphologically normal spermatozoa. 8 Though the diagnosis of male factor infertility appears straightforward by assessing these standard semen parameters, its etiology remains unclear. In fact, the underlying reasons for oligo-astheno-teratozoospermia (OAT) are still unknown and often progressive, rendering the effectiveness of conventional treatments aimed at sperm preparation extremely doubtful. Reports have drawn attention to the genetic etiology present in about half of the men experiencing loss of spermatozoal function. 9 However, the reasons for primary or secondary male infertility have yet to be fully elucidated and investigations have even been delayed because of advances in assisted reproduction. 10 The implementation of micromanipulation techniques in the past 20 years has made it possible to overcome male gamete production deficiencies and fertilization defects to allow infertile male partners to reproduce at rates that previously would have been deemed unachievable. Because of its wide utility and reliability, ICSI has been extensively adopted as the preferred insemination method in many circumstances According to the International Committee Monitoring Assisted Reproductive Technologies (ICMART), ICSI comprised 66% of all aspirations, a continued increase from 60.6% in 2004 and 62.9% in The proportion of ICSI procedures varied according to region: 96% in the Middle East, 81% in Latin America, 70% in North America, 76% in Europe and 56% in Australia and New Zealand. ICSI generated over 60% of the total ART children estimated at over 256,668 in At our center, there has been a steady and progressive increase in ICSI prevalence starting at 32% in 1993 rising to 50% in 1995, and reaching 73% by Our experience at Cornell parallels the widespread popularity of this technique. Since 1993, ICSI has been assimilated in over 37,000 assisted fertilization procedures consisting of 11,189 cycles with standard in vitro insemination and 26,429 with ICSI. ICSI has yielded comparable reproductive outcome in comparison to conventional IVF but is capable of consistently overcoming unforeseen sperm cell dysfunction. The fertilization after conventional insemination was 60.6% (74,741/123,316) versus 74.4% (161,842/217,449) after ICSI. However, with standard IVF, the two-pronucleated (2PN) formation rate is calculated over the total number of oocytes retrieved, so once corrected for ICSI, the fertilization is comparable between the two insemination methods (60.6 IVF vs. 58.5% ICSI). Clinical pregnancy rate as defined by the presence of a fetal heartbeat on ultrasound was 40.0% (4,473/11,189) for IVF compared with 45.7% (12,066/26,429) for ICSI. Thus far, we have 14,458 babies born by the two ART procedures, of which 9,284 of them were conceived with ICSI.

3 94 Intracytoplasmic Sperm Injection Palermo et al. Indications for Intracytoplasmic Sperm Injection IVF has become a well-established treatment for most types of infertility, including long-standing infertility due to tubal disease, endometriosis, unexplained infertility, and even some mild forms of male factor. 15 However, it soon became clear that certain couples with an extremely low sperm count, impaired motility, and poor morphology could not be helped by IVF alone. To tackle this problem, attempts to refine conventional IVF insemination practices, such as microdrop insemination, marginally helped conception in cases of sperm dysfunction. 16 Microdrop insemination is achieved by preparing 50- to 100-mL droplets of sperm suspension (concentration of 7, ,000 motile spermatozoa) that is covered with oil and oocytes incubated in the droplets until fertilization. Assisted fertilization procedures based on the micromanipulation of oocytes and spermatozoa, among them ICSI, proved to be a superior male infertility treatment modality. 5 Though ICSI requires only a spermatozoon with a functional genome and centrosome for the fertilization of the oocyte, indications for ICSI are not restricted to morphologically impaired spermatozoa; they also include low sperm count and impaired sperm kinetics. The ICSI procedure can be used successfully in patients with fertilization failure after conventional IVF, and also in patients with too few morphologically normal and progressive motile spermatozoa present in the ejaculate (<500,000 sperm cells). High fertilization and pregnancy rates are achieved when a viable spermatozoon is injected, independent of its characteristics. 17 Injection of exclusively immotile spermatozoa results in lower fertilization rate because it is impossible to distinguish the viable cells. 18 When only nonvital sperm cells are present in the ejaculate, the use of testicular spermatozoa is indicated. 19 Semen parameters, such as concentration, morphologic feature, and high titer of antisperm antibodies, 20 do not influence success rate. 21 Successful ICSI has also been described in patients with acrosomeless spermatozoa 22,23 with the exception of globozoospermia type I 21 in which procedures such as assisted oocyte activation needs to be used. 24 Any form of infertility due to obstruction of the excretory ducts can be treated by ICSI with spermatozoa microsurgically recovered from either the epididymis or the testis. 28,29 Obstructive azoospermia can result from congenital bilateral absence of the vas deferens, failed vasectomy reversal, or vasoepididymostomy. When no motile spermatozoa can be retrieved from the epididymis due to epididymal fibrosis, testicular spermatozoa can be isolated from a testicular biopsy specimen. Testicular biopsy has also proven useful in retrieving suitable spermatozoa in cases of nonobstructive azoospermia In patients with severe testicular dysfunction as a result of incomplete germ cell aplasia, Sertoli-cell only syndrome (SCOS), hypospermatogenesis, or incomplete maturation arrest 34 spermatozoa retrieval can be achieved by targeting specific foci in the germinal epithelium by micro- TESE. 35 In addition, cryopreservation of supernumerary spermatozoa recovered from the epididymis 36 or the testis 37 may avoid repeated surgery. Azoospermia caused by testicular failure can be treated by ICSI if enough spermatozoa can be retrieved in testicular tissue samples. Table 1 provides an overview of the current indications for ICSI. 38,39 ICSI cannot be applied when no sperm cells are available or when no mature oocytes are retrieved. ICSI failure rates range from 1 to 3% when a lower number of eggs are injected; these failures are caused by a combination of missed fertilization, abnormal fertilization including one and three pronuclei (diagynic), 39,40 or due to lysis of the oocytes. Rarer still, fertilization with ICSI may fail despite an adequate number of oocytes and sufficient spermatozoa available because of missed ooplasmic maturation due to nuclear cytoplasmic asynchrony. 7 A recurrent condition that is more difficult to address with ICSI is when few, motile spermatozoa with variable morphology are available. Our center treats severely oligospermic men with a concentration of /ml of spermatozoa. Between 1993 and 2013, 1,660 ICSI cycles using spermatozoa from these men had an average sperm concentration of /ml, with a motility of %, and normal morphology of only 1.0 2%. The fertilization in this group was 64.4% (10,131/15,738), resulting in an acceptable clinical pregnancy rate of 50.7% (842/1,660). Once a couple has decided to use their own gametes, even in the face of difficulties, one option is to attempt the ART cycle with ejaculated spermatozoa, possibly supported by some trials of specimen cryopreservation to safeguard the couple from rare but feasible occurrences of azoospermia at the time of egg retrieval. If the initial semen specimen examination in the Makler chamber yields no spermatozoa, a high-speed centrifugation is often able to find scarce cells. In 244 cycles, after high-speed centrifugation a mean density of /ml and motility of % was reached. In this cohort, we obtained a fertilization of 60.3% (1,500/2,488) and replacement of an average of 2.6 embryos per patient resulted in a satisfactory intrauterine pregnancy rate of 48.4% (118/224). We have recently published an in-depth analysis of cases where an extreme search is necessary to find scarce sperm cells, whether in the ejaculate or testicular biopsy. 41 When we examine the three different sperm sources for unselected cases encompassing all maternal ages ( Table 2), the ejaculated cohort displayed the highest fertilization rates despite containing the oldest women within the comparison (p <0.001). Epididymal spermatozoa achieved a somewhat lower fertilization but reported the highest clinical pregnancies defined by the presence of a least one fetal heartbeat. Cycles using testicular spermatozoa, even though this cohort contained the youngest female partners, attained the lowest fertilization rates (p ¼ ) and a somewhat lower pregnancy outcome (p ¼ ) ( Table 2). This analysis is purely academic, because the surgically retrieved spermatozoa address different clinical indications, but nevertheless is evidence of the ability of this powerful insemination method to utilize different spermatozoal maturational stages without meaningful outcome impairment. The peculiarity of the testicular sampling, particularly in relation to the unpredictability of sperm retrieval in nonobstructive azoospermic men (not to mention the limited

4 Intracytoplasmic Sperm Injection Palermo et al. 95 Table 1 Intracytoplasmic sperm injection indication Male Factor Ejaculated spermatozoa Oligozoospermia Asthenozoospermia Teratozoospermia Antisperm antibodies Fertility preservation Ejaculatory disorders Electroejaculation Retrograde ejaculation Acrosomeless spermatozoa Cryptozoospermia Surgically Retrieved Epididymal Obstructive Azoospermia Congenital bilateral absence of the vas deferens Young syndrome Failed vasoepididymostomy Failed vasovasostomy Testicular spermatozoa Necrozoospermia All indications for epididymal sperm Non-Obstructive Azoospermia Hypospermatogenesis Non-Male Factor Prior failed fertilization with IVF Oocyte dysmorphism Low number of oocytes Low oocyte maturity Cryopreserved oocytes In vitro maturation (IVM) PGD/PGS (FISH or PCR) HIV and Hepatitis C discordant couples Restrictive legislation Abbreviations: FISH, fluorescent in situ hybridization; HIV, human immunodeficiency virus; IVF, in vitro fertilization; PCR, polymerase chain reaction; PGD, preimplantation genetic diagnosis; PGS, preimplantation genetic screening. Table 2 Intracytoplasmic sperm injection outcome using ejaculated, epididymal, and testicular spermatozoa Sperm source Ejaculated Epididymal Testicular Maternal age a a a Cycles 24,021 1,013 1,395 Fertilization (%) 147,728/197,126 (74.9) b 6,871/9,649 (71.2) b 7,242/13,374 (54.1) b Clinical pregnancy (%) 10,986 (45.7) c 515 (50.8) c 565 (40.5) c a Student t-test, p < b,c χ 2,2 3, 2 df; effect of sperm source on fertilization and clinical pregnancy rates, p ¼ availability of skilled reproductive urologists), has supported the upsurge of cryopreservation of the testicular specimen. We therefore assessed, among the TESE cohort, the performance of spermatozoa used fresh (same day) or the day following the biopsy (433/1,031) in comparison to their frozen (132/364) counterparts. The clinical pregnancy rate appeared impaired by the cryostress on the male partner specimens (42.0% fresh vs. 36.3% frozen TESE; p ¼ 0.05). This analysis, however, included couples that failed the fresh cycle and returned to us utilizing the stored surplus specimen. To control for repeat-cycle bias, we then looked at cases that were treated using fresh and frozen testicular biopsy as first attempts ( Table 3). In spite of a slightly older couple age in the frozen group, the concentration and motility of the specimen was comparable albeit with a morphology that is obviously being unique and does not conform to the standard World Health Organization (WHO) criteria due to the direct gonadic retrieval. Both specimens required the use of a motility enhancer with the cryopreserved cohort in excess of the fresh. When fertilization characteristics were assessed, interestingly, the frozen cohort yielded higher numbers of zygotes than the fresh group (P ¼ ), leaving the

5 96 Intracytoplasmic Sperm Injection Palermo et al. Table 3 Semen parameters and fertilization of TESE cases at their first attempt using fresh versus frozen testicular biopsy TESE First attempt Fresh Frozen Cycles Male age (M SD y) Concentration (M 10 6 /ml SD) Motility (M% SD) Morphology (M% SD) 0 0 Pentoxifylline treatment (%) 399 (57.0) 98 (73.1) Female age (M SD y) Oocytes retrieved 9,259 1,485 MII injected 7,210 1,165 2PN (%) 3,910 (54.0) a 736 (63.2) a 1PN (%) 392 (5.4) 51 (4.4) 3PN (%) 179 (2.5) 37 (3.2) Abbreviations: M, mean; MII, metaphase II; 1PN, 1 pronucleus; 2/3PN, 2/3 pronuclei; SD, standard deviation; y, years. a χ 2,2 2, 1 df; effect of cryopreservation on normal fertilization rate, p ¼ abnormal fertilization similar, for example, one-pronucleus (1PN) and three-pronuclei (3PN) ( Table 3). The evolution of pregnancy was then evaluated. βhcg levels, clinical pregnancy rates, and delivery and ongoing pregnancies rates were comparable between the fresh and cryopreserved groups, when evaluated as a first attempt. The pregnancy loss rates were also equivalent at 3.1% for the fresh and 4.5% for the frozen counterpart ( Fig. 1). The Debate Fig. 1 Clinical pregnancy with fresh versus frozen testicular biopsy at first attempt. It would be naïve to think that ICSI is being used only when male factor infertility is present. According to 2012 SART registry, ICSI was used in over two-thirds of all ART cycles in the United States ( rptcsr_publicmultyear.aspx?clinicpkid¼0). The same database showed that male factor infertility was present in only 35% of cases. Indeed, the routine use of ICSI regardless of the infertility symptoms has been proposed. 42 The classic study to address this issue is a multicenter trial from the United Kingdom that randomized 415 couples with non male factor infertility treated by ICSI or conventional in vitro insemination. 43 Of note, fertilization rates per oocytes retrieved were 58% for IVF versus 47% for ICSI (p <0.0001) whereas the pregnancy rate were 33 versus 26%, respectively, without reaching statistical significance. However, the implantation rate was higher with conventional IVF than with ICSI (30 vs. 22%; p <0.05). In this study, ICSI resulted in a higher fertilization rate per metaphase II (MII) oocyte injected, because of a better detection of the oocyte maturational status after cumulus cells were removed. The authors concluded that ICSI requires significantly higher resources without offering any clinical advantage over conventional IVF for couples with non male factor infertility. Though it is unclear why the fertilization rate is within the low range of those achievable by ICSI in this study, the overall pregnancy rates for both groups were very low considering the extremely young female age (IVF years vs. ICSI years), suggesting a compounded unrecognized female factor. In addition, the fertilization rate per mature oocyte injected was 65% being at a very low end of an optimal ICSI fertilization. An outcome of complete fertilization failure following conventional IVF is a frustrating experience for normospermic patients as well as clinicians. This inability to generate zygotes affects between 5 and 15%. 44,45 Chances of recurrent fertilization failure in conventional IVF cycles may vary indicating that factors other than the male gamete may be responsible for the occurrence. This is not limited to ooplasmic inadequacy that can be corrected by tweaking the superovulation protocol and may also explain the paradoxical

6 Intracytoplasmic Sperm Injection Palermo et al. 97 higher fertilization rates observed with IVF over ICSI as the overnight exposure to the male gamete allows an unsuitable ootid to reach ooplasmic competence. 7,24,46,47 In one retrospective analysis of 555 couples with failed fertilization after IVF (n ¼ 388) or ICSI (n ¼ 167), couples who opted for ICSI on subsequent cycles achieved a clinical pregnancy rate of 24% compared with 22% for those who used standard insemination. Cases with failed fertilization with ICSI and subsequently inseminated by the same method generated a pregnancy rate of 22%. 48 The authors presented evidence that the fertilization failure was a result of suboptimal response to ovarian stimulation and not the actual insemination methods themselves. In fact, fertilization and cleavage rates with ICSI of similar number and quality embryos transferred in patients with previous failed/low fertilization with standard IVF had significant lower pregnancy rate than those with primary male factor (19.6 vs 33.5%, respectively). 46 The surprising outcome of this study after exclusion of male factor is puzzling and indicates an improper patient selection as argued by the extremely low pregnancy rate attained in such young female partners (IVF years vs. male factor years). In addition, the number of couples examined was extremely small and may not allow a meaningful conclusion to be drawn. Finally, the disclosed seasonal schedule, depicting a progressive increase in pregnancy rate, suggests that a procedural learning curve cannot be excluded. In a randomized study that compared IVF to ICSI in sibling oocytes, a 64.3% ICSI fertilization compared with 37.4% (p <0.007) with in vitro insemination utilizing 5,000 spermatozoa was observed. This rate could be increased to 59.6% by using more number of spermatozoa (20,000/oocyte), albeit inconsistently. 49 This correction, however, increased occurrence of polyspermia and the consequent epigenetic risk represented by the high reactive oxygen species (ROS) exposure due to the high sperm cell number. 50 In another study, ICSI performed superiorly compared with a prior IVF attempt that resulted in complete failed fertilization or low fertilization. 51 In this group, fertilization rate per oocyte retrieved was 48% for ICSI and 11% for conventional IVF in the complete fertilization failures. ICSI (60%) and IVF (22%) fertilization rates were similar in a group with impaired fertilization in prior cycles. This report suggests that in non male factor couples the use of ICSI will not grant higher pregnancy rates but may be justified to prevent unexpected fertilization failure or correct low fertilization rates. 52 Indeed, up to 30% of couples seeking fertility treatment are ultimately diagnosed with unexplained infertility. 53 It has been hypothesized that the mechanisms causing unexplained infertility include fertilization defects not directly attributable to the male gamete, as semen parameters are often within the normal limits in such cases. Though conventional in vitro insemination has been shown to be an effective therapy for most couples with unexplained infertility, fertilization failure may occur in a relevant portion of them. 49,54 Because it is difficult to predict which couples with unexplained infertility may experience fertilization failure, some physicians suggest the practice of splitting oocytes into ICSI and conventional insemination groups, despite an implied risk of generating a mixed population of embryos that, if simultaneously transferred, may lead to inconclusive results. A recent meta-analysis examined the fertilization rates per retrieved oocyte of couples with unexplained infertility in 11 randomized controlled studies. 55 In five of these studies, sibling oocytes were specifically assigned to ICSI or conventional IVF before assessment of maturity while no relevant information was presented in others. Fertilization per allocated oocyte was significantly higher for ICSI (relative risk [RR]: 1.43; 95% confidence interval [CI]: ) as confirmed by a higher RR of complete fertilization failure for all 11 trials included in the meta-analysis that was significantly higher with conventional IVF (RR: 8.22; 95% CI: ). In these trials, each patient served as her own control, thereby reducing the impact of potentially important confounders, such as maternal age, ovarian stimulation, oocyte and sperm quality, and laboratory conditions. Comparison of pregnancy and live birth rates between the two groups could not be made, because many patients received a combination of embryos derived from conventional insemination and ICSI. This confirms the point raised earlier: It is preferable to perform ICSI in these non male factor patients only after standard insemination has exhausted its therapeutic potential. It is important to note that the oocytes allocated to ICSI are routinely assessed for nuclear maturity and may contribute to a higher fertilization rate per oocyte compared with unselected oocytes undergoing conventional IVF. Although there are clues that ICSI may benefit some couples with unexplained infertility, more studies reporting pregnancy outcome are needed to clarify whether ICSI should be preferred to standard IVF for clear non male factor subfertility. 56,57 However, unexpected fertilization failure due to male gamete dysfunction not predicted by the standard semen analysis is of real concern in such cases. It has been hypothesized that ICSI might be a better method of fertilization than conventional IVF for patients with poor-quality oocytes, as determined by morphologic assessment. 58 Although on the question of whether the use of ICSI actually improves the pregnancy outcome in morphologically abnormal oocytes has not been convincingly answered, the early work of Van Blerkom 59 evidenced that oocytes with morphologic abnormalities (SER, vacuoles, dark center, etc.) fertilized poorly with standard in vitro insemination. It is still not clear whether suboptimal oocytes are linked to specific patient characteristics or were due to a woman s response to a particular superovulation regimen, for example, excessive LH. 60 The advantage of using ICSI may be twofold in these situations: ICSI confers the ability to read these oocytes following cumulus removal and specifically enhances fertilization chances by direct sperm injection. ICSI is commonly used in poor responders to ovarian stimulation with the idea of improving fertilization rates in the few oocytes that are available for fertilization. One prospective trial randomized 96 non male factor infertility cases with six or fewer retrieved oocytes were allocated to conventional IVF and ICSI. 61 The ICSI fertilization rate was 77.7% and 70.2% for IVF per inseminated oocyte yielding

7 98 Intracytoplasmic Sperm Injection Palermo et al. comparable quantity and quality of embryos, implantation, and pregnancy rates (17.3 vs. 21.1%, with ICSI and conventional insemination, respectively). This is surprising because of the relatively younger maternal age of 35.5 years for ICSI versus 36.7 years for IVF, with normal semen parameters. However, in these couples the low number of oocytes cumulus complexes may shroud immature oocytes, therefore depriving the ability for the reproductive specialist to tune up the stimulation protocol in poor responders or women of advanced reproductive age. Moreover, the unrecognized presence of dysfunctional gametes incapable of generating zygotes may prove disastrous for these couples. Another possible benefit for the approach of ICSI would be for the prevention of polyspermia. In fact, the reported incidence of triploidy in human embryos after conventional IVF ranges from 2 to 10%, with dispermy being the most common cause. 62 A retrospective analysis of 95 couples with 20% incidence of 3PN zygotes in their initial conventional IVF cycles followed by the use of ICSI in a subsequent attempt showed that after ICSI, the rate of normally fertilized zygotes (2PN) was corrected (65 vs. 34.1%) with a negligible occurrence of 3PN (5.0 vs. 33.9%). As expected, there was no difference in cleavage and quality of embryo derived from normal zygotes by the two insemination methods. Thus, ICSI generated a higher number of diploid zygotes without compromising embryo developmental characteristics and yielded a superior number of conceptuses. 63 In most cases the disproportionate occurrence of three pronuclei with standard insemination can be easily corrected by adjusting the number of spermatozoa used for insemination. Only in rare occasions is the block to polyspermic penetration altered at the zona level due to deficit in the ZP2 cleavage gene or in the defective appearance/release of the cortical granules. 64 More recently JUNO has been recognized as a block to polyspermia at the oolemma level that can only be overcome by ICSI. 65 We have actually observed this phenomenon in the rare occasion of a spermatozoon accidentally released in the perivitelline space just prior to successful injection. The second spermatozoon s participation in fertilization resulted in a dispermic 3PN zygote (Palermo personal communications). What may be worth mentioning is that the early methods to assisted fertilization methods such as SUZI, PZD and ZD were all plagued by an unacceptable rate of polyspermic fertilization at times as high as 40%. 66 On the other hand, ICSI itself can generate up to 3% of 3PN (diagynic) oocytes where the additional pronucleus is a decondensed and unextruded second polar body. This is often observed with the sudden membrane breakage pattern during oolemma penetration ANoNo ICSI has been proposed as a method to reinseminate oocytes that failed to fertilize following conventional IVF Though reinsemination of unfertilized oocytes 15 to 18 hours after the initial insemination has been performed with ICSI, normal fertilization rates after rescue ICSI remain relatively low and the generated embryos achieved low pregnancy rates. 74 In one study, rescue ICSI was performed on 439 oocytes from 54 couples with complete fertilization failure that resulted in a normal fertilization of 44% of the survived oocytes. Although 48 couples had at least one embryo transferred, only 8 pregnancies were reported (14.8%) with an implantation rate of 10.7%. 72 It is thought that oocyte quality diminished during the 24 hours after retrieval, and although some oocytes may still be fertilized, embryos derived from rescue ICSI procedures often arrest at early stages of development. Furthermore, high rates of polyploidy were reported in embryo fertilized by rescue ICSI, 72,74 and the proportion of abnormally fertilized oocytes seems to be a function of the length of time the oocyte is in culture following the initial failed fertilization assessment. 75 Moreover, reinsemination of oocytes by IVF 76,77 or by ICSI appears to generate more aneuploid embryos. 75 Considering the very low pregnancy rate, in addition to the potential genetic risks associated with the aging oocytes in culture, the authors agree with those who recommend against rescue ICSI. 74 The Answer Answering the question of whether or not ICSI should be performed in non male factor patients relies on the possibility of identifying spermatozoa with an intact fertilizing competence. This concealed potential of the male gamete can be overlooked by standard semen analysis, and the use of new methods capable of assessing sperm function would be revealing. 78 It has been claimed that the occurrence of DNA fragmentation in a sperm sample used through programmed intercourse, IUI, and seldom in IVF may have a small effect on fertility rates but large effects on embryo implantation and development. This relationship has not been confirmed when ICSI insemination is used. In fact, DNA fragmentation measured in a particular sample does not take into consideration whether the cells are motile and therefore functionally intact. 7,79 The unclear relationship between DNA integrity and pregnancy outcome with ICSI inseminations may be explained by the fact that only a motile spermatozoon is selected for injection whenever possible, 7 in addition to the other benefits inherent to this insemination method such as the limited exposure of the spermatozoon to laboratory and culture-induced damage. 50,80 82 One long unmet need in the diagnosis of male fertility is an indicator of sperm fertilizing ability. This would provide a much needed functional complement to descriptive parameters such as count, morphology, or DNA fragmentation. Defects in sperm function in the face of a normal semen analysis might account for idiopathic failure to fertilize during either intrauterine insemination or IVF. The ability to diagnose functional defects would therefore meet a great clinical need in choosing the most appropriate form of ART. Because of this need, many assays of sperm function have been suggested over the years (e.g., hamster zona penetration assays), but none are in common usage. A potentially promising diagnostic target relates to how sperm acquire the ability to fertilize. Ejaculated spermatozoa

8 Intracytoplasmic Sperm Injection Palermo et al. 99 are not immediately able to fertilize an egg. Rather, they must undergo a process of functional maturation known as capacitation. Currently, there are no sensitive and simple markers for capacitation that can be used in a clinical setting. For example, the appearance of protein tyrosine phosphorylation events during the process of capacitation has been described in many species including humans However, visualization of these events using polyacrylamide gel electrophoresis and immunoblotting can take upward of 48 hours to perform, making it unsuitable for clinical purposes. 86,87 Rather than targeting downstream processes in capacitation, recent attention has focused on early upstream events at the level of the sperm plasma membrane. It has been known for some time that one of the major requirements that the sperm must undergo in the female tract that enables them to fertilize is sterol efflux, or removal of cholesterol from the sperm plasma membrane. 88 Cholesterol and lipids such as the ganglioside G M1 have been shown to be organized in special microdomains on the plasma membrane, known as membrane rafts. 87,89 These organize a variety of signaling pathways making them attractive candidates to mediate sperm function. Interestingly, the patterns of localization of G M1 change in both mouse and bull sperm that have responded to stimuli and have capacitated. 90 These findings suggest that the pattern of G M1 localization can be used as a diagnostic marker to assess sperm function and therefore male reproductive fitness. Use of G M1 for this application makes physiologic sense. Recently, it was shown in mouse sperm that G M1 regulates the activity of an R-type calcium channel in sperm, triggering a transient calcium flux that is essential for the sperm to undergo acrosome exocytosis and fertilize. 91 If the localization behavior of G M1 couldbeusedasa biomarker for capacitation status, a test to detect it could be useful clinically, with the results helping clinicians to counsel their patients toward the most appropriate ART procedure for example, determining first whether to pursue IUI, and if that is not likely to be productive, whether to choose classical IVF or ICSI. Because of this potential, we have begun investigating G M1 localization in human sperm and our preliminary results are encouraging. 92,93 However, much work remains to be done to translate this prototype assay from a research laboratory setting to a commercially available assay that is appropriate for broad clinical use. 94 Particular challenges will involve finding a means to standardize the assay across clinics, validating the assay in a broader patient population including men of normal fertility, and establishing reference ranges that correspond to men of different fertility status. As with all the most recent attempts aimed at improving ART such as those achieved by time lapse imaging that use dynamic time-sensitive morphometric parameters to identify the ideal conceptus that will develop to blastocyst, 95 we have adopted genetic assessment and molecular markers to screen the proportion of competent spermatozoa. Presently, we are working on the creation of a new sperm imaging technology in which we capitalize on a sperm s innate ability to swim. This novel technique combines light microscopy with advanced computer vision algorithms to generate a three-dimensional view of the sperm s cellular surface. It is possible that in the near future, biomorphometric parameters based on digital imaging reconstruction can be linked to the spermatozoon s inner structures and corroborated by TEM, SEM, cytogenetic chromosomal mapping, and proteomic analysis. 96 With this new technology we would be able to gain information on the intimate structures of the cell not immediately discernible in vivo, enabling a more proper identification of a developmentally competent spermatozoon. Closure The question at hand to ICSI or not to ICSI is almost as thorny a question as asking when to perform a cesarean section. That surgical procedure was devised to overcome situation of cephalopelvic disproportion, extended labor, maternal exhaustion, to name just a few indications. It has garnered favor in countries such as Brazil, Italy, and some areas of the United States, in spite of the inherent risks linked to this unphysiologic delivery method. Similar to the C- section, that enables birth in the face of any adversity, ICSI certainly allows the highest chances of securing a fertilized conceptus independent of the availability of a large number of spermatozoa and of their functionality. As with the obstetric procedure, ICSI has specific indications and is to be used in case of failed fertilization or to prevent its failure with standard IVF. ICSI was developed in the late 1980s, when there was a fervid quest for a procedure to enhance gamete union and overcome the devastating effect of fertilization failure, which at that time affected up to 40% of the patients undergoing in vitro insemination. 1 Assisted fertilization procedures, among which ICSI surfaced clearly as the most effective, allowed treatment not only of men with dysfunctional spermatozoa but also individuals with severe oligospermia and even azoospermia at a time when those men were not even capable of biological fatherhood. ICSI requires technical skills that conventional insemination does not: It is performed out of the laminar flow hood, on a heated stage outside of the incubator, and requires an enzymatic/mechanical removal of the cumulus oophorus. It needs to be performed in a highly regulated laboratory environment (in fact, early ICSI adopters indirectly improved their pregnancy rates because of the required adjustments to more stringent laboratory conditions). Most importantly, ICSI must be performed in an expedited fashion. ICSI was developed to overcome the shortcoming of the spermatozoon and should be used only in such conditions or whenever there is a concern for a reduced ability of the male gamete to penetrate an oocyte. It is somewhat naïve to expect that ICSI in non male factor cases may actually increase pregnancy rates: In fact, study after study has shown that in the best scenarios it will only offer comparable results to conventional in vitro insemination. Indeed ICSI, because of its sensitivity, is more likely to be impaired by any fluctuation in laboratory procedural protocols or media or culture condition aberrations. Moreover, ICSI requires that the oocyte reaches nuclear and cytoplasmic maturity to be fertilized in a time-sensitive fashion once

9 100 Intracytoplasmic Sperm Injection Palermo et al. injection is performed. Conversely, oocytes that are nearing nuclear maturity or have just extruded the first polar body and require additional time to attain ooplasmic readiness may paradoxically benefit from standard in vitro insemination because of the overnight availability of the male gamete. There is a lack of randomized studies on the use of ICSI in non male factor couples because such studies are timeconsuming, expensive, and may not uphold the primum non nocere concept represented by the use of the least invasive procedure at first. For the reproductive specialist, the tendency to overuse ICSI is created by situations in which a failure of fertilization may prove devastating for the infertile woman or couple both economically and emotionally. On the other hand, the underuse of ICSI resulting in total fertilization failure may one day even be perceived as a liability issue. The solution to the quandary is in the ability to have information on the functional competence of the spermatozoon by measuring structural and dynamic functions of the acrosomal structures and in the interpretation of genetic or epigenetic markers that reveal the competence of the male gamete to perform as expected in physiologic insemination conditions. Acknowledgments We thank the clinicians, embryologists, andrologists, and scientists of the Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine and Urology Department, Weill Cornell Medical College. We are grateful to Stephanie Cheung, Tyler Cozzubbo, Theodore Paniza, and Laura Park for their assistance in the laboratory, to Dr. Brian Levine for his attempts at generating a 3D model of the spermatozoon, to Dr. Alexander Travis for his valuable input on the GM 1 assay, and to Daniel Pepper for editing the manuscript. References 1 Cohen J, Edwards RG, Fehilly CB, et al. Treatment of male infertility by in vitro fertilization: factors affecting fertilization and pregnancy. Acta Eur Fertil 1984;15(6): Palermo G, Van Steirteghem A. Enhancement of acrosome reaction and subzonal insemination of a single spermatozoon in mouse eggs. Mol Reprod Dev 1991;30(4): Palermo G, Joris H, Devroey P, Van Steirteghem AC. Induction of acrosome reaction in human spermatozoa used for subzonal insemination. Hum Reprod 1992;7(2): Palermo G, Joris H, Derde MP, Camus M, Devroey P, Van Steirteghem A. Sperm characteristics and outcome of human assisted fertilization by subzonal insemination and intracytoplasmic sperm injection. Fertil Steril 1993;59(4): Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 1992;340(8810): 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(6): Palermo GD, Neri QV, Monahan D, Kocent J, Rosenwaks Z. Development and current applications of assisted fertilization. Fertil Steril 2012;97(2): WHO Laboratory Manual for the Examination and Processing of Human Semen, 5th ed. Cambridge: Cambridge University Press; Reijo R, Lee TY, Salo P, et al. Diverse spermatogenic defects in humans caused by Y chromosome deletions encompassing a novel RNA-binding protein gene. Nat Genet 1995;10(4): Turek PJ. Practical approaches to the diagnosis and management of male infertility. Nat Clin Pract Urol 2005;2(5): Aboulghar MA, Mansour RT, Serour GI, Sattar MA, Amin YM. Intracytoplasmic sperm injection and conventional in vitro fertilization for sibling oocytes in cases of unexplained infertility and borderline semen. J Assist Reprod Genet 1996;13(1): Fishel S, Aslam I, Lisi F, et al. Should ICSI be the treatment of choice for all cases of in-vitro conception? Hum Reprod 2000;15(6): Nyboe Andersen A, Carlsen E, Loft A. Trends in the use of intracytoplasmatic sperm injection marked variability between countries. Hum Reprod Update 2008;14(6): Mansour R, Ishihara O, Adamson GD, et al. International Committee for Monitoring Assisted Reproductive Technologies world report: Assisted Reproductive Technology Hum Reprod 2014;29(7): Steptoe PC, Edwards RG. Birth after the reimplantation of a human embryo. Lancet 1978;2(8085): Trounson AO. The choice of the most appropriate microfertilization technique for human male factor infertility. Reprod Fertil Dev 1994;6(1): Van Steirteghem AC, Nagy Z, Joris H, et al. High fertilization and implantation rates after intracytoplasmic sperm injection. Hum Reprod 1993;8(7): Nagy Z, Liu J, Cecile J, Silber S, Devroey P, Van Steirteghem A. Using ejaculated, fresh, and frozen-thawed epididymal and testicular spermatozoa gives rise to comparable results after intracytoplasmic sperm injection. Fertil Steril 1995;63(4): Tournaye H, Liu J, Nagy Z, Verheyen G, Van Steirteghem A, Devroey P. The use of testicular sperm for intracytoplasmic sperm injection in patients with necrozoospermia. Fertil Steril 1996;66(2): Nagy ZP, Staessen C, Liu J, Joris H, Devroey P, Van Steirteghem AC. Prospective, auto-controlled study on reinsemination of failedfertilized oocytes by intracytoplasmic sperm injection. Fertil Steril 1995;64(6): Liu J, Nagy Z, Joris H, Tournaye H, Devroey P, Van Steirteghem A. Successful fertilization and establishment of pregnancies after intracytoplasmic sperm injection in patients with globozoospermia. Hum Reprod 1995;10(3): Lundin K, Sjögren A, Nilsson L, Hamberger L. Fertilization and pregnancy after intracytoplasmic microinjection of acrosomeless spermatozoa. Fertil Steril 1994;62(6): Bourne H, Richings N, Harari O, et al. The use of intracytoplasmic sperm injection for the treatment of severe and extreme male infertility. Reprod Fertil Dev 1995;7(2): Neri QV, Lee B, Rosenwaks Z, Machaca K, Palermo GD. Understanding fertilization through intracytoplasmic sperm injection (ICSI). Cell Calcium 2014;55(1): Silber SJ, Nagy ZP, Liu J, Godoy H, Devroey P, Van Steirteghem AC. Conventional in-vitro fertilization versus intracytoplasmic sperm injection for patients requiring microsurgical sperm aspiration. Hum Reprod 1994;9(9): Tournaye H, Devroey P, Liu J, Nagy Z, Lissens W, Van Steirteghem A. Microsurgical epididymal sperm aspiration and intracytoplasmic sperm injection: a new effective approach to infertility as a result of congenital bilateral absence of the vas deferens. Fertil Steril 1994;61(6): Mansour RT, Aboulghar MA, Serour GI, Fahmi I, Ramzy AM, Amin Y. Intracytoplasmic sperm injection using microsurgically retrieved epididymal and testicular sperm. Fertil Steril 1996;65(3): Devroey P, Liu J, Nagy Z, Tournaye H, Silber SJ, Van Steirteghem AC. Normal fertilization of human oocytes after testicular sperm extraction and intracytoplasmic sperm injection. Fertil Steril 1994;62(3):