ICSI for non-male factor: do we practice what we preach?

Abstract

Since its introduction in 1992, intracytoplasmic sperm injection (ICSI) has revolutionized the treatment of infertility due to severe male factor. Over the last three decades, the use of ICSI for non-male factor has increased dramatically, despite guidelines to the contrary from professional societies. Excessive utilization of ICSI is primarily due to an irrational fear of total fertilization failure, which is at odds with rational evidence to support its use.

In this month’s issue, Haas et al. report the results of single-center prospective randomized trial to evaluate the role of intracytoplasmic sperm injection (ICSI) in women aged 39 and older undergoing treatment for non-male factor indications [1]. For each study participant, the investigators randomized oocytes from one ovary to ICSI and those from the contralateral side to conventional fertilization, in order for patients to serve as their own controls. The fertilization rate and the mean number of total and top-quality embryos were similar between groups, suggesting that ICSI does not lead to significantly improved outcomes in this group of patients.

The introduction of ICSI in 1992 was a major milestone in the history of assisted reproductive technology (ART), revolutionizing the treatment of male factor infertility [2]. For the spectrum of various male factor conditions, the benefit of ICSI is undisputed [3]. However, the dramatic rise in the use of ICSI has been most pronounced in cycles done for non-male factor indications, increasing from one out of every six US cycles in 1996 to two-thirds of cycles in 2012 [4].

For some non-male factor indications, there is solid evidence to support the use of ICSI. In the setting of previously frozen oocytes, ICSI can overcome cryopreservation-induced hardening of the zona pellucida [5]. The same principle applies to the context of in vitro maturation of oocytes, as hardening of the zona has also been observed in this setting [6]. With the increasing use of pre-implantation genetic testing (PGT), DNA contamination from sperm present in biopsy samples obtained after conventional fertilization has been cited as an indication for the use of ICSI. However, the advent of next-generation sequencing (NGS) has largely rendered this argument invalid, and recent evidence suggests that sperm DNA fails to amplify under conditions used to process trophectoderm samples in PGT, making the risk of paternal cell contamination negligible [7]. Likewise, there is currently no evidence to support the use of ICSI in the setting of unexplained infertility, low oocyte yield, advanced maternal age, or as the sole fertilization method for all cases [3]. In the context of prior absent or low fertilization using conventional fertilization, ICSI can lead to improved fertilization rates in an ensuing treatment attempt [8]. However, the number of all-comer non-male factor infertility couples needed to have undergone ICSI to prevent one case of failure of (conventional) fertilization has been estimated to be 33 [3, 9]. In other words, 32 of 33 couples would have to undergo ICSI unnecessarily to prevent one extremely disappointing outcome for the 33^rd couple. Choosing ICSI as the sole fertilization method would be justifiable if there was no downside to it compared to conventional fertilization. However, the ICSI procedure increases the workload of the IVF laboratory and the cost to the patient. Evidence from large population cohort studies also suggests that the use of ICSI is associated with an increased risk of birth defects compared to conventional IVF—although it remains unclear whether this finding is due to the intervention or the patient population receiving it [3, 10].

The evidence demonstrating the absence of a benefit of ICSI for non-male factor infertility is not tenuous. A 2019 European multicenter retrospective analysis by Drakopoulos et al. in 4891 patients with non-male factor infertility revealed no advantage of ICSI over conventional fertilization in “poor” (1–3 oocytes), “suboptimal” (4–9 oocytes), “normal” (10–15 oocytes), and “high” responders (> 15 oocytes) [11]. The authors concluded that the number of oocytes retrieved should not be a factor in the decision regarding the method of fertilization. A single-center prospective pilot study in 30 couples with non-male factor infertility undergoing preimplantation genetic testing for aneuploidies (PGT-A) found similar euploid rates in blastocysts from sibling oocytes randomized to conventional IVF versus ICSI [7]. The results of a larger prospective multi-center RCT to assess the effectiveness of ICSI versus conventional IVF in couples with non-male factor infertility will provide more insight into this question, with live birth rates as the primary endpoint (NCT03428919) [12].

Why is there a discrepancy between evidence-based recommendations for the use of ICSI on the one hand and actual practice on the other? Anyone who has ever had to tell a hopeful couple that none of their 14 retrieved oocytes actually fertilized knows the answer. Clinical decisions are not only made rationally. Our field abounds with examples of deviations from the recommended practice due to emotional and often illogical closely held beliefs. Denial, selective consideration of the literature, and cognitive dissonance can be observed on both sides of the argument in a variety of controversial academic discussions in our specialty, including those on the appropriate use of elective single embryo transfer (eSET), PGT-A, or the endometrial receptivity assay (ERA). The irrational use of tests and procedures in medicine is so widespread that the “choosing wisely” initiative was created by the American Board of Internal Medicine (ABIM) to facilitate conversations between clinicians and patients with the goal to promote care which is “supported by evidence, not duplicative of other tests or procedures already received, free from harm, and truly necessary” [13]. The American Society for Reproductive Medicine (ASRM) and the American College of Obstetricians and Gynecologists (ACOG) are specialty society partners of the “choosing wisely” initiative.

Attempts to explain deviations from the standard of care or recommendations from professional societies lead to the study of human behavior. In psychology, “cognitive dissonance is mental stress or discomfort experienced by an individual who holds two or more contradictory beliefs. The brain strives for consistency” [14]. Examples of cognitive dissonance in everyday life and from history are plentiful: the “meat paradox” is the psychological conflict between people’s dietary preference for meat and their moral response to “animal suffering” [15]. Aesop’s sixth-century Greek fable “the Fox and the Grapes” is about a fox who cannot reach grapes high up on a vine and then postulates without evidence that the grapes are sour and that he did not want them anyways (even though he reached for them). Reconciling evidence demonstrating an absence of benefit of ICSI for non-male factor infertility with the fear of the uncommon but devastating event of total fertilization failure can lead to cognitive dissonance in the fertility provider’s brain striving for consistency. Like the fox in Aesop’s fable, we may generate plausible explanations as to why we do not follow evidence-based guidelines, with statements such as “In our personal experience ICSI leads to better outcomes” or “Patients are asking for it.” The force of habit and the comfort of a steady routine in the IVF laboratory are likely contributing factors to the overuse of ICSI—it may be simpler and less prone to human error to handle all cases the same than to individualize fertilization methods.

In this context, the metaphorical grapes are not actually too high—for most clinics, it would be feasible to restrict the use of ICSI to indicated cases—it would just take a different kind of effort to reach for the grapes and pick them. Using conventional IVF when indicated requires extra time spent counseling before and after the cycle. Prior to the cycle, the lack of benefit of ICSI for most non-male factor cycles and the small risk of failed fertilization need to be reviewed. After the cycle, fertility providers must be prepared to spend extra time in uncomfortable conversations with the few unlucky patients with the devastating outcome of total fertilization failure. As Keating and Palermo point out in their commentary on a new meta-analysis concluding that the use of ICSI is not indicated for non-male factor infertility [16], “ICSI is here to stay” [17]. It just needs to be utilized for the indication it was developed for, and not as a first-line fertilization method for cases with presumably healthy male and female gametes.

In a field where new interventions and add-ons emerge on an almost daily basis, it will be difficult for providers and patients to accept that sometimes “less is more,” even though fewer interventions would likely result in higher cost-effectiveness. The interplay between guidelines and actual practice will continue to evolve. While practical aspects and human factors need to be considered, our specialty should continue to leap for the grapes, rather than concluding they are sour because they seem out of reach.