The need to improve patient care through discriminate use of intracytoplasmic sperm injection (ICSI) and improved understanding of spermatozoa, oocyte and embryo biology

Abstract

The discovery that intracytoplasmic sperm injection (ICSI) can overcome many types of infertility is one of the greatest advancements in the history of reproductive medicine. This technology has benefited millions of infertile couples (particularly those with male factor infertility) who, as a result, were able to conceive children using defective gametes previously considered unsuitable for conventional in vitro fertilization (IVF). Nevertheless, an important recent study by Boulet et al. (2015), supported by prior studies, highlights the growing trend of the overuse of ICSI for reasons other than male factor infertility and the apparent lack of benefit to the patient in such cases. In this editorial, we draw attention to the need for a renewed emphasis on the use of ICSI for the proper clinical indications, the potential risks and disadvantages of ICSI compared with conventional IVF, and the need for the development and implementation of diagnostic assays to assist in identifying patients who would realize the most benefit from ICSI. In 2001, Bhattacharya et al. (2001) published results of the first randomized, controlled trial of ICSI vs. conventional IVF in non-male factor patients and demonstrated that the implantation and pregnancy rates were lower using ICSI than conventional insemination. In 2011, Nangia et al. (2011) evaluated 465 046 assisted reproductive technology (ART) cycles reported in the Society for Assisted Reproductive Technology Clinic Outcome Reporting System database from 2004 to 2008 and compared outcomes for ICSI and non-ICSI cycles for patients with no reported male factor contribution to infertility. The data showed a significant decrease in clinical pregnancy rates and an increase in fetal loss for non-male factor patients receiving ICSI therapy compared with conventional IVF. Nevertheless, the increasingly common trend of use of ICSI for broader or poorly defined indications continues, particularly for patients with reduced oocyte yield upon controlled ovarian stimulation and patients with advanced maternal age. This increased use of ICSI continues despite studies that refute any benefit in such cases (Jain & Gupta, 2007; Centers for Disease Control & Prevention ASfRM, Society for Assisted Reproductive Technology, 2013). Furthermore, some IVF clinics now use ICSI as the standard insemination technique for all IVF patients, regardless of clinical phenotype. Boulet et al. (2015) analyzed the largest dataset yet, consisting of 1 395 634 fresh IVF cycles reported to the US National Assisted Reproductive Technology Surveillance System between 1996 and 2012. During that period, the incidence of ICSI usage for cycles without male factor infertility increased from 15.4 to 66.9% (p < 0.001), with significant and similar increases in usage observed for patients with unexplained infertility, patients with a maternal age >38, patients with <5 oocytes retrieved, patients with two or more previously unsuccessful ART cycles, patients undergoing pre-implantation genetic diagnosis (PGD) and patients with no reported specific indication. In all, ICSI was performed in 51.2% of cases without male factor infertility. Those ICSI cycles had a lower implantation, pregnancy, and live birth rate compared with non-ICSI cycles. Evidently, those patients who had ICSI without a male factor could be negatively biased in terms of their prognosis, beyond what could be accounted for in multivariable models including patient and clinical characteristics (Boulet et al., 2015). While no dataset is perfect, it is important to note that the results are consistent with the prior large database analyses reported by Jain & Gupta (2007) and Nangia et al. (2011). The Practice Committees of the American Society of Reproductive Medicine and the Society for Assisted Reproductive Technology reached similar conclusions and state in their guidelines that the data do not support the use of ICSI for advanced maternal age, low oocyte yield or unexplained infertility (Practice Committees of the American Society for Reproductive Medicine & Society for Assisted Reproductive Technology, 2012). Similarly, the British National Institute of Clinical Excellence (NICE) guideline is to use ICSI ‘in cases with severe deficits in sperm quality, and to consider ICSI for couples in whom a previous IVF treatment cycle has resulted in failed or very poor fertilization’ (http://www.nice.org.uk, 2015). The use of ICSI may be partially attributed to non-medical factors. Jain & Gupta (2007) showed that the use of ICSI was higher in states with mandated insurance coverage than those without. Nyboe Andersen et al. (2008) show that ICSI usage varies widely between European countries, despite its increased cost compared with conventional IVF, although no clear medical necessity was demonstrated and no association between usage and outcome success was observed. The latest available European data (Kupka et al., 2014) show that ICSI was carried out in 83.3–89.4% of all stimulated cycles in Italy and Spain, and 98% of IVF cycles in Turkey (Ferraretti et al., 2012). Thus, certain countries have almost abandoned the use of routine IVF. In contrast, IVF remains the method used in around 50% of all cycles in the Netherlands, United Kingdom and the Nordic countries (Kupka et al., 2014). Collectively, these studies highlight the possibility that non-medical factors influence the use of ICSI. Such factors may even have potential financial benefits to the clinic. Clinicians may argue that they do consider medical factors in selecting ICSI for non-male factor infertility, as ICSI use is prevalent in women where the risk of fertilization failure following IVF is expected to be increased: unexplained infertility, poor ovarian response to stimulation, advanced maternal age and repeated unsuccessful ART cycles (Boulet et al., 2015). However, complete fertilization failure is difficult to foresee in cases of non-male factor infertility, as neither duration of infertility or earlier in vivo conceptions predict fertilization failure in IVF (Krog et al., 2015). We believe that clinical decision-making to use ICSI is highly influenced by the concept of ‘better safe than sorry’, when informing patients that ‘everything was performed’ to enhance fertilization and the generation of embryos for transfer. This mantra may underlie the continually increasing rise observed in ICSI use, but unfortunately it suggests a lack of confidence in implementing guidelines based on the best available evidence. Intracytoplasmic sperm injection is associated with a small, but significant increase in genetic and epigenetic abnormalities in the offspring (Bonduelle et al., 2002; Zhu et al., 2006; Amor & Halliday, 2008; Gjerris et al., 2008; Davies et al., 2012; Whitelaw et al., 2014). This elevated risk is likely because of inherent abnormalities in the gametes of patients with severe spermatogenesis or oogenesis defects; however, it is important to note that prior studies clearly demonstrate that post-ICSI fertilization dynamics are different than what is observed with conventional IVF. For example, Hewitson et al. (2002) showed that ICSI was associated with altered sperm chromatin decondensation, retention of the perinuclear theca, and altered pronuclear formation, all of which may affect cell cycle dynamics and epigenetic remodeling of the male pronucleus. Furthermore, in animal and human studies ICSI is associated with altered early embryonic transcription and decreased number of cells in the inner cell mass (Giritharan et al., 2010). These studies do not imply that ICSI is unsafe, but rather emphasize that ICSI and conventional fertilization are likely different on subcellular and molecular levels, and that those differences may have the potential to alter important cell functions, including, potentially, epigenetic alterations that may affect the offspring long-term (Jenkins & Carrell, 2012). Those risks are not yet proven or disproven in patients undergoing ICSI with normal sperm quality, but should elicit caution in clinical practice. The molecular defects underlying the majority of male infertility remain largely unknown or undiagnosed. Indeed, in many IVF clinics the spermatozoa is only considered to provide a vehicle to supply the paternal DNA to the oocyte. Some may consider fertilization an indicator of normal spermatozoa. However, it is apparent that poor spermatogenesis may portend other health anomalies or risks. In a landmark paper by Honig et al. (1994), significant medical pathologies were uncovered in 1.1% of the 1236 clinic patients who presented with male infertility. Ten of these patients (0.8%) had tumors. Although, perhaps, it was not surprising that six men had testis cancer given their spermatogenic deficiency, three of the 10 men had brain tumors and one a spinal cord tumor. Five men had various endocrine anomalies and several others ejaculatory dysfunction because of mesonephric duct anomalies. Indeed, a comprehensive male infertility evaluation identified a number of significant, previously undiagnosed medical pathologies, some of which were potentially treatable (Honig et al., 1994). More recent analysis of non-obstructive azoospermic patients revealed that they have a 2.9-fold increased risk of cancer (Eisenberg et al., 2013). Hotaling & Walsh (2009) showed the risk of testis cancer was nearly threefold higher in 22 562 partners of infertile couples compared with normal men. This patient cohort was also 2.6 times more likely to be diagnosed with high-grade prostate cancer (Hotaling & Walsh, 2009; Walsh et al., 2010). A similar study of more than 32 000 men over an approximately 30-year period revealed a 1.6-increased risk of testis cancer, cancers of the peritoneum and other digestive organs in men with lower sperm count (Jacobsen et al., 2000). These findings point to the need for urologic consultation and in depth analysis when an infertile couple seeks treatment at an ART clinic, particularly in patients with apparent male factor infertility. Another of the unfortunate consequences of the successes of ICSI is a decreased emphasis on the study of sperm function. Tests of sperm function are rarely if ever ordered today (optimized sperm penetration assay, hemizona-type assays, acrosome reaction assays, etc.), even though clinical insights and benefits have been demonstrated in many cases (Gvakharia et al., 2000; Oehninger et al., 2014). Sperm fluorescent in situ hybridization analysis (Ramasamy et al., 2014) is rarely performed despite its ability to diagnosis meiotic defects resulting sperm aneuploidy in men with oligoathenoteratozoospermia, severe oligospermia, spermatogenic failure and men with generally normal semen parameters who are partners of women with recurrent pregnancy loss all of whom are commonly candidates for ICSI–IVF. Furthermore, clinical translation of research results showing that specific genetic causes of globozoospermia/round headed spermatozoa acrosome absence, misplacement or atrophy (such as DPY19L2 deletion or mutation, SPATA16 mutation, PICK1 mutations) portend poor ICSI-IVF outcomes has been slow or non-existent (Harbuz et al., 2011; Koscinski et al., 2011) and has not led to routine patient diagnosis even though this genetic information aids in prediction of ICSI outcome. The same is true for patients with immotile spermatozoa because of primary ciliary dyskinesia with mutations in ZMYND10, HYDIN who require ICSI to achieve a pregnancy. These mutations result in various phenotypic anomalies such as situs inversus/laterality defects and airway disease in addition to infertility (Olbrich et al., 2012; Moore et al., 2013; Zariwala et al., 2013; Ramasamy et al., 2014). These studies highlight the apparent disregard of existing studies on sperm function and biology that ultimately may help in the selection of patients for which ICSI is truly warranted, as well as maximizing IVF outcome and safety. In summary, ICSI is invaluable in treating patients with male factor infertility, however, such patients should be evaluated by an andrologist to assure that ICSI is indeed necessary and that broader health concerns are addressed. ICSI may also be indicated in other situations, including the use of cryopreserved oocytes, and in conjunction with PGD, however, the data clearly show no benefit in routine use of ICSI for other patients.